| Human Molecular Genetics | Pages |
Mutations in Emery-Dreifuss muscular dystrophy and their effects on emerin protein expression
Introduction
Results
Discussion
Materials And Methods
Clinical features of patients
PCR primers
PCR and DNA sequencing
SDS-PAGE and western blotting
Subcellular fractionation
Immunofluorescence microscopy
Acknowledgements
References
Mutations in Emery-Dreifuss muscular dystrophy and their effects on emerin protein expression
Seventeen families with Emery-Dreifuss muscular dystrophy (EDMD) have been studied both by DNA sequencing and by emerin protein expression. Fourteen had mutations in the X-linked emerin gene, while three showed evidence of autosomal inheritance. Twelve of the 14 emerin mutations caused early termination of translation. An in-frame deletion of six amino acids from the C-terminal transmembrane helix caused almost complete absence of emerin from muscle with no localization to the nuclear membrane, although mRNA levels were normal. This shows that mutant emerin proteins are unstable if they are unable to integrate into a membrane. A 22 bp deletion in the promoter region was expected to result in reduced emerin production, but normal amounts of emerin of normal size were found in leucocytes and lymphoblastoid cell lines. This shows that DNA analysis is necessary to exclude emerin mutations in suspected X-linked EDMD. Emerin levels in female carriers often deviated from the expected 50% and this was due, in at least two families, to skewed emerin mRNA expression from the normal and mutated alleles. In one family with a novel deletion of the last three exons of the emerin gene, a carrier had a cardiomyopathy and very low emerin levels (<5% of normal) due to skewed X-inactivation. In the three autosomal cases of EDMD, emerin was normal on western blots of blood cells, which suggests that autosomal EDMD is not caused by indirect reduction of emerin levels.
INTRODUCTION
Emery-Dreifuss muscular dystrophy (EDMD) is usually inherited as an X-linked recessive disorder, although an autosomal dominant form has also been described (1). There are clear clinical criteria for diagnosis of Emery-Dreifuss disease in a family (2), and these features are similar for both X-linked and autosomal forms. They include (i) early contractures of the elbows, Achilles tendons and spine; (ii) slowly progressive muscle wasting, mainly of the upper arm and lower leg (calf hypertrophy as in Duchenne/Becker dystrophies usually excludes the diagnosis); and (iii) cardiac conduction defects. The X-linked disorder is caused by mutations in the gene for emerin, a 254 amino acid, serine-rich protein which is ubiquitously expressed in most tissues and which contains two regions of homology with thymopoietins (3). In 1996, emerin was shown to be absent in several X-linked EDMD patients studied and to be localized at the nuclear rim in muscle and several other tissues (4,5). Emerin belongs to a family of type II integral membrane proteins which are anchored to the inner nuclear membrane via hydrophobic tails, with the remainder of the molecule projecting into the nucleoplasm (5). This family includes lamina-associated protein 2 (LAP2; now known to be [beta]-thymopoietin) (6,7) and the lamin B receptor (8). The presence of emerin at the nuclear membrane in many non-muscle tissues has enabled diagnostic tests for emerin abnormalities to be performed on blood samples (leucocytes) or skin biopsies without the necessity of an invasive muscle biopsy (9,10). Recently, Cartegni et al. (19) have suggested that cardiac conduction defects in EDMD might be explained by an additional localization of emerin at intercalated disks in the heart.
Although a large number of different mutations in the emerin gene which cause X-linked EDMD have been identified (11), the effects of the mutation on emerin protein production have been reported in only a few of these so far (4,5,9,10). We now describe the effects on emerin protein expression of 14 different mutations in the emerin gene. Although a few missense mutations and in-frame deletions have been identified, most mutations in the emerin gene are point mutations or frameshifts which are predicted to cause early termination of translation. In theory, a truncated emerin molecule could be produced in such cases, even though loss of the C-terminal anchor would prevent localization to the nuclear membrane, but they have rarely been detected.
Table 1
| Family | Mutation | Emerin protein |
| F1/2708 | nt953 del CAGT Early stop | |
| Proband 12 years (01) | mut + | absent |
| Mother 36 years (03) | mut + | normal* |
| Male cousin 44 years | mut + | absent |
| F2/2268 | nt1675 del TCCG Early stop | |
| Proband 22 years | mut + | absent |
| Brother 19 years | mut + | absent |
| Mother 53 years | mut + | 50% |
| Sister 20 years (08) | mut + | 50% |
| Sister 23 years | mut - | normal |
| Aunt | mut - | n.d. |
| Aunt | mut - | n.d. |
| Male cousin 27 years (21) | mut + | absent |
| Sister 22 years | mut - | normal |
| Sister | mut - | normal |
| Male cousin 16 years (11) | mut + | absent |
| Mother (13) | mut + | normal* |
| F3/2408 | nt386 del C Early stop | |
| Proband | mut + | absent |
| F4/1579 (Fig. 4) | del exons 4-6 Early stop | |
| Proband 15 years (01) | mut + | absent |
| Mother 41 years (03)(symptomatic)> | mut + | <5%* |
| F6/2197 | nt112. T→G Loss of start codon | |
| Proband 25 years (01) | mut + | absent |
| Mother 39 years (03) | mut + | normal* |
| Sister 27 years (04) | mut + | normal* |
| Sister 28 years | mut - | normal |
| F7/3078 | nt193 del G Splice site donor | |
| Proband | n.d. | n.d. |
| Mother 57 years (03) | mut + | 50% |
| Sister 25 years (04) | mut + | near-normal* |
| L1/3193 | nt363 C→T Early stop | |
| Proband | mut + | absent (blood and skin) (7) |
| Brother | absent (blood and skin) | |
| Mother | 28% (blood) 35% (skin)* | |
| Aunt | reduced (blood) | |
| Great-aunt | reduced (blood) | |
| L2/3063 | nt909 T→G Early stop | |
| Proband | mut + | absent (blood) |
| Mother | mut + | 50% (skin) |
| Aunt | mut - | normal (skin) |
| Male cousin | mut + | absent (blood) |
Table 1. Continued
| Family | Mutation | Emerin protein |
| Brother | mut - | normal (blood) |
| Mother | mut + | n.d. |
| L4/3162 | nt919 G->T Early stop | |
| Proband | mut + | absent (skin and blood) |
| Mother | 50% (skin) | |
| Female cousin | Normal (blood)* | |
| Son | absent (blood) | |
| L5/3266 (Fig. 2) | del18bp (nt1759/1763 to 1776/1780) In-frame removal of codons 235-240 or 236-241 |
|
| Proband | mut + | <5% (muscle blot) |
| Sister | mut + | n.d. |
| Sister | mut + | n.d. |
| Sister | mut - | n.d. |
| S1/2864 (Fig. 5) | nt994 G->A Splice-site donor lost in intron 4 | |
| Proband 13 years | mut + | absent (muscle and blood) |
| Brother 5 years | mut + | absent (muscle and blood) |
| Mother | mut + | 70-95% (skin) normal (blood)* |
| G-EMD-1 | nt908 del AT Early stop (14) | |
| Proband (2026) | mut+ | absent |
| G-EMD-3 (Fig. 3) | nt -19 to -40 del22 bp in promoter region (14) | |
| Proband (1677) | mut+ | normal (LCL and blood) |
| Mother | mut+ | present |
| Aunt | mut+ | present |
| Aunt | mut+ | present |
| Grandmother | mut+ | present |
| G-EMD-4 | nt631 delTCTAC Early stop | |
| Proband (3089) | mut+ | absent |
| Brother (3091) | mut+ | absent |
| Brother | mut - | n.d. |
| Brother | mut - | n.d. |
| Sister | mut - | n.d. |
| Mother (3086) | mut - (mosaic) | normal |
This has led to the suggestion that emerins which lack the hydrophobic C-terminal helix are unstable in the cell because they are unable to integrate into membranes (4). Alternative possibilities are that mutated emerin mRNAs are unstable [transcript reduction model (12)] or that altered emerins are unstable because they are unable to fold correctly.
Figure 1. Western blots of emerin in samples from EDMD families and sporadic cases. Tissue samples were lymphoblastoid cell lines (LCLs) in (a) and (b: lanes 8-14) and blood leucocytes in (b: lanes 1-7). See text and tables for full explanation of each result. The numbers above each lane correspond to the family numbers in Tables 1 and 2. Equal amounts of protein were loaded onto each lane of 4-12% acrylamide gradient SDS gels, and the corresponding Coomassie blue staining is also shown in (a). The mAbs used to detect emerin were MANEM5 in (a) and MANEM1 in (b) (5); the latter mAb cross-reacts with some higher molecular weight non-emerin bands in LCLs, providing a useful internal control for equal protein loading. The lower molecular weight bands in (b) are emerin degradation products. After treatment with peroxidase-labelled second antibody (a and b: lanes 8-14) or a peroxidase-biotin-avidin system (Vectastain in b: lanes 1-7), blots were developed by the chemiluminescence method. In (b: lane 6), biotinylated markers (Sigma) are 29, 40, 60 and 97 kDa. Fifty percent of normal emerin expression is expected in female carriers if X-inactivation is random, but deviation from this proportion has been reported (9). We have therefore investigated emerin expression in large numbers of EDMD family members for the first time. Mutations in genes other than emerin which are responsible for the rare autosomal forms of EDMD have not yet been identified. It is possible that autosomal EDMD mutations affect emerin levels or emerin function indirectly since the pathologies of X-linked and autosomal forms are indistinguishable. This possibility has now been studied in known and suspected autosomal EDMD families and in >20 sporadic cases. Table 1 summarizes the results of mutation analysis and emerin protein analysis in 14 EDMD families. Some of the western blot results on which the table is based are shown in Figure 1. . In 12 families, the emerin mutation was identified in this study by DNA sequencing, while in two additional families with published mutations, emerin protein analysis has now been performed. Of these 14 mutations, five are base substitutions (three creating early stop codons, one removing a splice site and one removing the ATG codon for initiation of translation) and six are small deletions of 1-5 bp (five frameshift deletions and one removing a splice site). The three mutations to stop codons and the five frameshift deletions are clearly predicted to cause early termination of mRNA translation and production of emerin molecules with various C-terminal truncations, all lacking the hydrophobic, transmembrane sequence near the C-terminus (Table 1). Family G-EMD-4 appears to be a case of germline mosaicism, since the mutation was not detected in leucocytes from the mother but two of her sons had the mutation. The ATG initiation codon knockout (Table 1: F6/2197) would also be predicted to prevent emerin production. Material to perform emerin protein analysis was available from the probands in nine of these 11 families, and in every case emerin was completely absent from western blots of lymphoblastoid cells lines (LCLs), leucocytes or skin biopsies (Fig. 1). Since the monoclonal antibodies (mAbs) recognize epitopes encoded near the beginning of exon 4 of the emerin gene (A. Pereboev and G.E. Morris, unpublished observations), smaller protein bands corresponding to truncated emerins would be detectable in some cases (e.g. families F1, F2, L5 and S1 in Table 1), but they were not detected on the western blots. This suggests that truncated emerin molecules are unstable in the cell either because they are unable to locate to the nuclear membrane correctly or because they are unable to fold correctly. Three EDMD families with larger deletions were particularly informative. Family L5/3266 has only the second in-frame deletion in the emerin gene yet reported. This 18 bp deletion is predicted to remove ~30% (six amino acids) of the transmembrane helix, and emerin was absent from the nuclear membrane in a muscle biopsy from the proband, with no evidence for relocation elsewhere (Fig. 2a). Western blots (Fig. 2b) show that emerin is still present in the muscle, but in greatly reduced amounts. RT-PCR on the muscle biopsy, using glycerol kinase as an internal control (15), revealed normal emerin mRNA levels (Fig. 2c), ruling out the possibility that mRNA instability might cause the reduced emerin levels. This result shows that a functional transmembrane helix is required for emerin stability in the cell and that the length of the transmembrane sequence is critical for correct insertion of emerin into the nuclear membrane. Figure 2. Emerin protein and mRNA expression in muscle from an EDMD patient with a six amino acid deletion in the transmembrane hydrophobic helix (family L5/3266). (a) Emerin is not detectable at the nuclear membrane in a muscle biopsy section. The mAb is MANEM5. (b) Emerin is only just detectable on the western blot of a muscle biopsy (<5% of the control biopsy). The mAb is MANEM1 and the blot was developed with Vectastain and diaminobenzidine substrate. P, patient; C, control. (c) Evidence that emerin mRNA levels are normal in this EDMD patient. C, control cDNA; E, EDMD patient cDNA. Lane 1, amplification of the proximal region (exons 1-5, 30 bp) of EDMD cDNA; lane 2, amplification of the distal region (exons 5 and 6, 451 bp) of EDMD cDNA; lane 3, the proximal region of EDMD cDNA was co-amplified with glycerol kinase (GK) cDNA (596 bp) as internal control; lane 4, the distal region of EDMD cDNA was co-amplified with GK cDNA; lane 5, amplification of GK cDNA alone. Densitometry showed that the amount of internal control GK transcript is 30% of the EDMD transcript in both control and patient (reactions 3 and 4). As expected, the distal EDMD PCR product appears slightly shorter in the patient whose genomic DNA is 18 bp deleted in exon 6 (reactions 2 and 4). The complete coding region of the EMD cDNA was amplified using primers P/C-P/1 or P/5-PD and the GK cDNA was amplified using primers S3B-S4. Family G-EMD-3 has a 22 bp deletion in the promoter region. The proband was a sporadic case, but segregation of the mutation in the pedigree (Fig. 3a) is at least compatible with X-linked inheritance. This mutation would be expected to have a significant effect on emerin mRNA and protein levels. Surprisingly, however, emerin protein levels were normal on western blots of both LCLs and leucocytes from this family (Fig. 3b). No other tissues are available for study. No other mutations have been found in the complete gene sequence, and the deletion is unlikely to be a polymorphism since it has not been observed previously in sequence data from >50 X-chromosomes. Subcellular fractionation of the LCL gave similar results to a control LCL, nearly all the emerin being found in the nuclear fraction in both cases (Fig. 3c). The patient (G1677; aged 7 years) showed progressive muscle weakness from age 4, elevated creatine kinase (CK) (50-100 times normal), early contractures of Achilles tendons with Achillotomy performed at ages 4 and 6, limited flexion of the upper spine, severe myopathic biopsy at age 4 with normal dystrophin expression and no major dystrophin deletion detectable by Southern blot. Apart from sinus arrhythmia, there was no cardiac conduction defect or cardiomyopathy, but these do not usually develop in EDMD before the second decade (1). Although several instances of very large deletions which encompass the whole emerin gene have been recorded, family F4/1579 is the first case of a large deletion within the emerin gene. The last three exons which encode nearly two-thirds of the emerin protein are deleted. At age 14, the proband had a particularly severe heart block, with permanent auricular paralysis requiring a pacemaker; emerin is absent. The carrier mother is symptomatic and has a pacemaker, while a maternal grandmother also displayed EDMD features and died of acute pulmonary oedema (16). Emerin protein levels were greatly reduced in LCLs from the mother to <5% of control levels (Fig. 1a, lane 16), which suggests that the emerin gene mutation is responsible for the clinical features, although the 3[prime]-end of the deletion has not yet been determined (flanking markers ST14 and F8C are present). Skewed X-inactivation has been demonstrated in the mother by methylation-sensitive restriction enzyme digestion of the androgen receptor locus (results not shown) and can explain why emerin levels are much lower than the 50% predicted for random X-inactivation. The effects of splice site mutations on emerin production are difficult to predict because of the possible use of alternative splice sites. Family F7/3078 (Fig. 4a) has a deletion of G in the splice site donor of intron 1 (Fig. 4b). RT-PCR has demonstrated the use of a cryptic splice site in exon 1 which results in an effective frameshift deletion in the mRNA of 29 bp (Fig. 4d). Two carriers showed both normal and deleted RT-PCR products, but the mother expressed mostly the deleted mRNA while the sister expressed mostly the normal mRNA (Fig. 4c); this may be due to differences in X-inactivation. The difference is reflected in emerin protein levels which are clearly reduced in the mother (3078-03) but near-normal in the sister (3078-04) (Fig. 1a, lanes 13 and 7). Figure 4. Mutation analysis in family F7/3078: mutation affecting a splicing site. (a) Pedigree of the family: affected males are shaded and obligate carriers are indicated by a central dot. (b) ABI electropherograms of sequence portions in the EDMD genomic DNA of control (upper) and subject 4 in the pedigree (lower). The arrowhead indicates the deleted G. The sequence of subject 4, heterozygous for the deletion, shows the shift downstream from the deleted G. EDMD genomic DNA was amplified using primers P/F-ED2.R. (c) RT-PCR products synthesized from RNAs of LCL of control (C) and obligate carriers (subjects 4 and 7 in the pedigree). Both carriers show a shorter band in addition to the normal one (430 bp). The deleted transcript is expressed preferentially in the mother's cDNA, whereas the normal transcript is expressed more in that of the daughter. Direct sequencing of the shortened cDNA characterized a 29 bp deletion upstream from the deleted G, starting with a cryptic GT donor site (data not shown). The exon 1-5 coding region of the EDMD cDNA was amplified using primers P/C-P/1. (d) Schematic representation of the abnormal splicing of intron 1. Upper: exons 1 and 2 of genomic DNA. The deleted G is pointed out on the sequence by the black arrowhead. The abnormally spliced part of exon 1 resulting in a deletion at the transcript level is shaded. Lower: deleted transcript. The deleted sequence of exon 1 is underlined and the cryptic splice site donor GT is in bold. Three families which meet all the clinical criteria for EDMD (except X-linkage), but have no detectable mutation, were found to have normal levels of emerin protein on western blots of LCLs or leucocytes (Table 2 and Fig. 1). These families may have autosomally inherited forms of EDMD, and this is supported by pedigree data in family H-EMD-1 in which females are also affected. Finally, we examined >20 sporadic cases displaying the Emery-Dreifuss syndrome. Some of these had been shown to have no mutation in the emerin gene by DNA sequencing, and this was confirmed by normal emerin levels in LCLs and leucocytes. Some sporadic cases were identified first by emerin protein analysis but, in some cases, an emerin mutation subsequently was identified in both proband and carriers, and these cases are included in Table 1. Table 2 We have shown previously, in one EDMD family, that female carriers can be identified by reduced levels of emerin on western blots and a mosaic pattern of emerin expression by immunofluorescence microscopy of skin sections (9), and we wished to determine whether these methods are generally applicable. In one large EDMD family, LCLs were available from six female members for western blot studies. DNA analysis showed that three of them did not have the mutation and they had normal emerin levels. Of the three that carried the emerin mutation, two had reduced emerin levels (e.g. Fig. 1a, lane 6) but one had apparently normal levels (Fig. 1a, lane 12). In other families, we also found instances where the western blot did not distinguish carrier from non-carrier females reliably (Fig. 1and Table 1). In small skin biopsies, deviation from the expected 50:50 distribution of emerin-positive and emerin-negative nuclei in carriers could be due either to skewed X-inactivation or to sampling variation. Figure 5. shows that sampling variation is significant when taking skin biopsies. Two separate biopsies were taken from this carrier; one showed 95% emerin-positive nuclei (Fig. 5a and b), which might be questionable as evidence for carrier status, but the second showed 70% emerin-positive nuclei (Fig. 5c), which is quite conclusive. This shows that a mosaic pattern of emerin expression in skin sections is a more reliable indicator of carrier status than western blots since carriers can still be identified even when 95% of the nuclei are emerin-positive (this would not be detectable as a difference in intensity on western blots). Figure 5. Mosaic expression of emerin in skin biopsies from an X-linked EDMD carrier. Immunofluorescence microscopy of emerin in one skin biopsy from the female carrier in family S1/2864 is shown in (a) and (b), while a second skin biopsy taken from a different part of the arm at the same time is shown in (c). Nearly all the nuclei in (a) are emerin-positive, except for a few nuclei at the top (shown at higher power in b). The second biopsy had ~70% emerin-positive nuclei overall. The mAb used was MANEM5. So far, most cases of EDMD with an established emerin gene mutation have been found to have no detectable emerin on western blots (this study; 4,5,9,10). The absence of emerin on western blots will therefore identify most cases of X-linked EDMD. This study, however, has revealed one case of EDMD with a 22 bp deletion in the emerin gene promoter in which normal amounts of normal size emerin are still produced in leucocytes and LCLs. DNA analysis will continue to be necessary, therefore, to exclude emerin mutations completely in suspected EDMD. It is also conceivable that mutations in the nucleoplasmic domain of emerin occur which abolish emerin function but allow it to locate to the nuclear membrane in normal quantities via its intact C-terminus. Mora et al. (10) have described an EDMD patient producing only slightly reduced amounts of normal size emerin with a Gln133->His mutation. A five amino acid deletion in exon 4 and several missense point mutations in this region have been identified in EDMD families (11), but their effects on emerin protein have not yet been reported. Most mutations in the emerin gene result in early termination of mRNA translation, and there are three possible reasons for the failure to detect truncated emerin proteins in many of these cases; the mRNA may be unstable (transcript reduction model), the protein may be unstable because it cannot fold correctly or the protein may be unstable because it lacks the C-terminal helix and cannot integrate into a membrane. The data from family L5/3266 in Figure 2. strongly support this last idea since they show that a nearly full-length emerin with only a slight shortening of the transmembrane helix is unstable in the cell. mRNA levels were normal, and the altered region is unlikely to affect protein folding in the nucleoplasmic domain since it normally is buried within the membrane. Recently, Cartegni et al. (17) have shown that the last 28 amino acids of emerin are sufficient for nuclear membrane localization in transfected COS cells. They also described an early termination frameshift mutation in which a shorter emerin is still made and locates to the nuclear membrane. In this case, however, the frameshift causes the synthesis of a novel C-terminal sequence and this, by chance, includes 21 hydrophobic amino acids which presumably are responsible for the localization. This is consistent with the results presented here. Emerin levels in this patient were greatly reduced, however, which implies that either the specific sequence of the hydrophobic region or the nature of the flanking sequences has some effect on emerin stability. It is not clear how the 22 bp deletion in the emerin gene promoter can bring about EDMD since emerin levels were normal in LCLs and its nuclear localization was normal (Figs 1. and 3. ). It is possible that emerin levels are more affected in heart and skeletal muscle than in LCLs, but we have not been able to check this since muscle biopsies are not available. If there are differences in promoter function between muscle and blood cells, then experimental studies of the role of this deleted region of the promoter could be a key to understanding the molecular pathogenesis of EDMD. The evidence that the emerin mutation is responsible in some way for the disease appears quite strong, though the possibility of a second mutation in an autosomal EDMD gene in the patient cannot be ruled out. The extent to which emerin levels must be reduced to produce EDMD is not clear. Some asymptomatic carriers have emerin levels reduced by >50% without showing symptoms (9). The required reduction may be different in patients, where emerin is reduced in all nuclei, compared with carriers, where the overall emerin level is determined by the relative proportions of emerin-negative and emerin-positive nuclei. The carrier mother in family F4/1579 was symptomatic, requiring a pacemaker at age 23, and had only 5% of normal emerin levels. We have shown that the wide range of emerin expression observed in female carriers, deviating considerably from the 50% predicted by random X-inactivation (18), is due to skewed expression of emerin mRNA from the normal and mutated allele in two different families. In one family, F4/1579, we have shown directly that this is due to skewed X-inactivation. One consequence of this is that detection of carriers by reduced emerin levels on western blots is unreliable. We have shown, however, that this problem can be largely avoided by immunolocalization of emerin on skin biopsies, since even a small percentage of emerin-negative nuclei can be detected when emerin levels appear normal on blots (cf. family S1/2864; Table 1 and Fig. 5). It has been known for some time that an autosomal form of EDMD exists, possibly more than one form (1). Other members of the emerin-related nuclear protein family should be considered as possible candidate genes; these include LAP2 on chromosome 12 and lamin B receptor on chromosome 1, and even nuclear lamins themselves. Three families and >20 sporadics without a mutation in the emerin gene were found to have normal levels of emerin protein, showing that the Emery-Dreifuss syndrome was not caused in these cases by indirect effects on emerin levels. Family H-EMD-1 shows clear evidence of autosomal inheritance with affected females, and the other two families are consistent with autosomal inheritance. It is accepted that all the features used to identify EDMD families may not be evident in any one individual, especially in cases that present early. The emerin western blot (or immunohistochemistry, if muscle or skin biopsy is available) should be used routinely for rapid screening of patients with problem myopathies, since EDMD cases with quite unusual clinical presentations have been identified in this way [e.g. L2/3063 in Table 1 (13)]. The cause of most of the sporadic EDMD cases in this study is unclear, since only a few had emerin mutations or abnormal emerin protein. If autosomal mutations were responsible, one would expect a higher proportion of affected females, and these are rare, both in general (1) and in our study, where we have seen only one affected female with no emerin mutation (Table 2). All patients investigated followed the consensus diagnostic criteria for EDMD (2). X-linked inheritance was proven in all except sporadic cases and those families specifically discussed in the Results section. EDMD (3,19,20) primers used were: P/F, gcggccgtgacgcgacaacg; ED2.R, ctccccgcgtcccaggctgtc; P/C, cgcctgagcccgcacccgc; P/D, cccactgctaaggcagtcagc; P5, atcaccaggtgcatgatgac; P/1, gtcatcatgcacctggtgatg. P/F-ED2.R (528 bp = around normal size), P/1 and P/C (430 bp = normal size), P/5 and P/D (451 bp = normal size). Glycerol kinase (15) primers were: S3B, ctttttgggactattgattcatgg; S4, caataaggtgcatataaccccg; S3B-S4 (596 bp = normal size). Reverse transcription from total RNA was performed using random hexamers and Moloney murine leukaemia virus reverse transcriptase (Gibco BRL). PCR amplification from cDNA was performed in a 50 µl reaction with 50 pM of each dNTP, 10 pmol of each primer, 5 mM MgCl2, 10 mM Tris, 50 mM KCl, 1% dimethylsulfoxide (DMSO) and 0.4 U of Taq polymerase (Boehringer) for 30 cycles (1 min at 94°C, 2 min at 72°C, 1 min at 56.5°C with 7 min of final extension). For amplification from genomic DNA, 1 µM of each dNTP was used and the PCR cycle was 30 s at 96°C, 2 min at 72°C, 1 min 30 s at 65°C and 7 min of final extension. The PCR products were run in a 5% acrylamide gel (19:1) for size analysis or in a 1% Nusieve-1% agarose gel for purification. Fifty ng of the purified PCR product were sequenced directly using the Taq DyeDeoxy terminator cycle sequencing kit as recommended by the supplier (Perkin-Elmer/ABI) using primer P/C. SDS-PAGE and western blotting were carried out essentially as described elsewhere (21). Antibody-reacting bands were visualized following development with a biotin-avidin detection system for mouse immunoglobulin (Vectastain kit; Vector Laboratories) and diaminobenzidine or by a chemiluminescence method using kits from Pierce or Amersham International. LCL cell pellets were homogenized in ice-cold RSB buffer (10 mM NaCl/1.5 mM MgCl2/10 mM Tris-HCl, pH 7.5) using a Dounce hand homogenizer. All steps were carried out at 4°C. The homogenate was centrifuged at 1000 g for 10 min. The pellet was extracted with 1% Triton X-100/10 mM Tris-HCl, pH 7.5 and centrifuged at 100 000 g for 30 min. Skin biopsy samples were taken from the forearm and were stored and transported at -70°C. Unfixed, frozen sections of human and rabbit tissues (6 microns) were mounted on glass slides and stored at -70°C. Fluorescein isothiocyanate (FITC)-labelled anti-mouse IgG (or biotinylated anti-mouse Ig and Texas red-avidin) was used to detect bound antibody as described previously (22). Twenty µl of mAb MANEM5 [diluted 1 + 3 in phosphate-buffered saline (PBS)] was placed on the surface of each tissue section for 30 min at room temperature. After washing twice in PBS (5 min), FITC-labelled rabbit anti-(mouse Ig) [diluted 1 + 20 in PBS containing 1% horse serum, 1% fetal calf serum and 0.1% bovine serum albumin (BSA)] was added in a dark-box for 30 min. After washing three times in PBS, the section was mounted in glycerol-based medium and viewed with a Leica DMRB photomicroscope with epifluorescence optics. We thank R. Warzok (Institute for Pathology, Greifswald) for immunohistochemical analysis of G-1677, Fiona Wilkinson and Tracy Fitzgerald (NEWI) for technical assistance, B. Schlosshauer and I. Jenz (EMAU) for technical assistance with LCLs and DNA sequencing and P. Warrot, C. Giraudet and F. Fraisse (LBGM) for technical assistance with LCLs. This work was supported by grants from the British Heart Foundation (G.E.M.), the Muscular Dystrophy Group of Great Britain (G.E.M.), Medical Research Council (C.A.S.) and Deutsche Forschungsgemeinschaft (We1470 to M.W.). For referral of EDMD case material, we are grateful to D. Hilton-Jones and W. Squire (Oxford), J. Colomer (Barcelona), D. Amsellem (Besancon), V. Layet (Le Havre), F. Pouget (Marseille), F. Corbery (Elbeuf) and N. Romero, V. Cormier-Daire and I.A. Urtizberea (Paris).
RESULTS
Figure 3. Subcellular fractionation of LCLs from an EDMD patient with a 22 bp deletion in the emerin promoter (family G-EMD-3). (a) Family pedigree. The mutation was detected by PCR amplification of the promoter region as described in (14) and agarose gel electrophoresis. Carriers showed two bands corresponding to the normal and mutated alleles (insets). The affected male (G1677) has a shorter PCR band from the mutated allele, whereas two unaffected males (G3935 and G1675) have the full-length PCR product from their normal allele. (b) There is no reduction in emerin levels in LCLs from this patient (lane 2) compared with a control LCL (lane 1). Equal amounts of total protein were loaded for both patient and control LCL extracts. (c) The total extracts (lanes 1 and 6) of LCLs from patient and a control were separated into nuclear pellet (lanes 2 and 7) and supernatant (lanes 3 and 8) by a low-speed centrifugation. The nuclear pellet was extracted with 1% Triton X-100 to yield a Triton-resistant pellet (lanes 4 and 9) and a Triton-soluble supernatant (lanes 5 and 10). All the emerin detected was in the nuclear fraction. Nuclear emerin was almost completely insoluble in Triton X-100 in both cases.
Mutation
Emerin protein
(a) Families
M-EMD-1
Affected male
none
present in leucocytes and LCL
Affected brother
none
present in leucocytes and LCL
Unaffected mother
none
present in leucocytes and LCL
Unaffected sister
none
present in leucocytes and LCL
H-EMD-1
Affected male (2737)
none
present in LCL
Affected female (2573)
none
present in LCL
Unaffected male
none
present in LCL
Unaffected female (2574)
none
present in LCL
D-EMD-3
Affected male
none
present in leucocytes
Unaffected male
none
present in leucocytes
Unaffected male
none
present in leucocytes
Unaffected female
none
present in leucocytes
Unaffected female
none
present in leucocytes
(b) Sporadics (all males)
G-3655
n.d.
absent in leucocytes
G-3604
none
present in LCL
G-3315
none
present in LCL
G-3313
none
present in LCL
L3/5032
none
present in skin and leucocytes
Eleven cases
none
present in LCLs
Three cases
none
present in leucocytes
DISCUSSION
MATERIALS AND METHODS
Clinical features of patients
PCR primers
PCR and DNA sequencing
SDS-PAGE and western blotting
Subcellular fractionation
Immunofluorescence microscopy
ACKNOWLEDGEMENTS
REFERENCES
This page is run by Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, as part of the OUP Journals
Comments and feedback: www-admin{at}oup.co.uk
Last modification: 4 Apr 1998
Copyright© Oxford University Press, 1998.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  S. Ura, Y. K. Hayashi, K. Goto, M. N. Astejada, T. Murakami, M. Nagato, S. Ohta, Y. Daimon, H. Takekawa, K. Hirata, et al. Limb-Girdle Muscular Dystrophy Due to Emerin Gene Mutations Arch Neurol, July 1, 2007; 64(7): 1038 - 1041. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. B. Yaou, A. Toutain, T. Arimura, L. Demay, C. Massart, C. Peccate, A. Muchir, S. Llense, N. Deburgrave, F. Leturcq, et al. Multitissular involvement in a family with LMNA and EMD mutations: Role of digenic mechanism? Neurology, May 29, 2007; 68(22): 1883 - 1894. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. L. V. Broers, F. C. S. Ramaekers, G. Bonne, R. B. Yaou, and C. J. Hutchison Nuclear lamins: laminopathies and their role in premature ageing. Physiol Rev, July 1, 2006; 86(3): 967 - 1008. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Ozawa, Y. K. Hayashi, M. Ogawa, R. Kurokawa, H. Matsumoto, S. Noguchi, I. Nonaka, and I. Nishino Emerin-Lacking Mice Show Minimal Motor and Cardiac Dysfunctions with Nuclear-Associated Vacuoles Am. J. Pathol., March 1, 2006; 168(3): 907 - 917. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Sakata, M. Shimizu, H. Ino, M. Yamaguchi, H. Terai, N. Fujino, K. Hayashi, T. Kaneda, M. Inoue, Y. Oda, et al. High Incidence of Sudden Cardiac Death With Conduction Disturbances and Atrial Cardiomyopathy Caused by a Nonsense Mutation in the STA Gene Circulation, June 28, 2005; 111(25): 3352 - 3358. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Tulac, C. Dosiou, E. Suchanek, and L. C. Giudice Silencing lamin A/C in human endometrial stromal cells: a model to investigate endometrial gene function and regulation Mol. Hum. Reprod., October 1, 2004; 10(10): 705 - 711. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 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, and C. J. Hutchison Both emerin and lamin C depend on lamin A for localization at the nuclear envelope J. Cell Sci., March 9, 2002; 114(14): 2577 - 2590. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. A. L. Fairley, A. Riddell, J. A. Ellis, and J. Kendrick-Jones The cell cycle dependent mislocalisation of emerin may contribute to the Emery-Dreifuss muscular dystrophy phenotype J. Cell Sci., January 15, 2002; 115(2): 341 - 354. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Boyle, S. Gilchrist, J. M. Bridger, N. L. Mahy, J. A. Ellis, and W. A. Bickmore The spatial organization of human chromosomes within the nuclei of normal and emerin-mutant cells Hum. Mol. Genet., February 1, 2001; 10(3): 211 - 219. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. E. Morris and S. Manilal Heart to heart: from nuclearproteins to Emery-Dreifuss muscular dystrophy Hum. Mol. Genet., September 1, 1999; 8(10): 1847 - 1851. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Fairley, J Kendrick-Jones, and J. Ellis The Emery-Dreifuss muscular dystrophy phenotype arises from aberrant targeting and binding of emerin at the inner nuclear membrane J. Cell Sci., January 8, 1999; 112(15): 2571 - 2582. [Abstract] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||