Introduction

Emery-Dreifuss muscular dystrophy (EDMD) is a rare inherited disorder presenting in childhood or adolescence with relatively benign neuromuscular features and potentially fatal cardiac involvement. It is characterized by the triad of (1) early contractures of the Achilles tendons, elbows, and posterior cervical muscles; (2) slowly progressive muscle weakness and wasting, with humeroperoneal distribution in the early stages; and (3) cardiomyopathy with conduction defects. The disease is transmitted usually as an X-linked recessive trait, rarely as an autosomal dominant (AD) trait, and exceptionally as an autosomal recessive (AR) or X-linked dominant trait. First clinically defined almost half a century ago, the progress made in the last 20 years, with several genes discovered to be associated with the EDMD phenotype, has provided tremendous insight into its pathogenesis. Strikingly, all these genes encode nuclear envelope proteins, so that it may be appropriate to classify EDMD under the rubric “nuclear envelopathies.”

Historical Background

In the early 1900s, Cestan and LeJonne and later Schenk and Mathias reported patients with muscular dystrophy and early contractures. In 1961, Dreifuss and Hogan described a large family from Virginia with “an unusual type of benign X-linked muscular dystrophy.” Contractures and cardiac abnormalities were not noted at the time, and the authors proposed that the slowly and rapidly progressive muscular dystrophies be grouped under an all-embracing term such as X-chromosomal or Duchenne-type muscular dystrophy. A reassessment of the same family a few years later by Emery and Dreifuss made it clear that they were facing a unique nosological entity. In particular, they drew attention to the presence of contractures and the absence of pseudohypertrophy as being different from Becker muscular dystrophy. Cardiac involvement was considered an essential part of the clinical picture. Thus, the symptoms of EDMD had been delineated. The report of Emery and Dreifuss was followed by a period of confusion, when cases with similar phenotypes were published under different designations. Rowland and coworkers suggested the use of the eponym Emery-Dreifuss muscular dystrophy for all these cases, based on the remarkably constant features of the clinical phenotype.

Several linkage studies pointed the localization of the gene to the long arm of the X chromosome. The gene responsible for X chromosome-linked EDMD (X-EDMD), STA or EMD , and the gene product, emerin, were identified in 1994. While the X-linked form was clinically well established in the literature, several reports of families showing male-to-male inheritance made it clear that an autosomal dominant form (AD-EDMD) with essentially the same clinical features also existed. The gene for the AD and AR forms, LMNA on 1q21, encoding lamin A/C, was identified in 1999.

About 50% of EDMD patients do not carry mutations in EMD or LMNA , suggesting the existence of other causative genes. Recently, mutations in SYNE1 and SYNE2 , encoding nesprin 1 and nesprin 2 proteins, were also found to be associated with the AD form of EDMD. The latest additions to the genes causing EDMD were FHL1 encoding four-and-a-half-LIM protein 1 (FHL1) and TMEM43 encoding LUMA.

Molecular Biology

Pathogenesis

The proteins that play a key role in the pathogenesis of EDMD are located in the nucleus, in contrast to the proteins involved in most other muscular dystrophies that are associated with the sarcolemma.

A definition of the NE and a brief summary of the localization and interrelationships of the nuclear proteins involved in the pathogenesis of EDMD are necessary to have some idea of the complex. The NE acts as a barrier separating the nucleus from the cytoplasm and is essential for the maintenance of the nuclear shape and its integrity. It consists of two membranes (inner and outer) and a dense network of nuclear intermediate filaments formed by A-type and B-type lamins. Lamins A and C are the major isoforms of A-type lamins.

The NE contains over 100 different proteins, the functions of most of which still remain unexplored. Three groups of NE proteins relevant to the pathogenesis of EDMD are LEM-domain proteins (emerin and others), SUN-domain proteins (SUN1 and SUN2), and KASH-domain proteins (nesprins 1, 2, 3, and others). Emerin is an integral membrane protein localizing to the inner membrane. It binds to lamins (most importantly to the A-type lamins) and to BAF (barrier-to-autointegration factor), which is a highly conserved protein essential for chromosome segregation, cell cycle progression, and postmitotic nuclear assembly. This trio (emerin-lamins-BAF) forms a major component of the NE-associated nucleoskeletal structure known as “nuclear lamina.” In addition to their role in maintaining the mechanical stability of the NE throughout the phases of the cell cycle, emerin and lamins interact with many proteins that regulate DNA synthesis, chromatin organization, gene transcription, and cell differentiation.

Emerin also binds directly to SUN-domain and KASH-domain proteins. SUN-domain proteins, which span the inner nuclear membrane, bridge to KASH-domain proteins, which span the outer nuclear membrane. SUN- and KASH-domain proteins form the LINC-complex (linker of nucleoskeleton and cytoskeleton), which is proposed to form a mechanical link between the nucleoskeleton and cytoskeleton.

As can be seen, the proteins causing EDMD, emerin, lamin A/C, nesprin 1, and nesprin 2, interact with each other. LUMA is also a binding partner of emerin. The interactions of FHL-1 with NE proteins are not yet known. FHL1 is localized to the sarcomere, the sarcolemma, and the nucleus, and has important roles in maintaining the stability of the sarcolemma, myofibrillar assembly, and transcriptional regulation. Absence or reduced level of FHL1 protein may cause delayed myotube formation.

Disrupted function of these proteins is the key pathophysiological mechanism in EDMD. Accumulating data suggest that EDMD might be caused by the uncoupling of the nucleoskeleton and cytoskeleton because of damaged emerin-nesprin-lamin interactions. However, how the mutations cause disease is still far from being elucidated. It is particularly puzzling that these ubiquitous proteins lead to tissue-specific disorders. Proposed hypotheses on pathogenesis include NE defects affecting nuclear stiffness and increased susceptibility to mechanical stress, NE defects as determinants of altered nucleocytoplasmic interplay, altered cell cycle control, alterations of the nuclear morphology affecting chromatin rearrangements, disruption of the process of skeletal muscle regeneration, altered DNA repair due to oxidative stress, and increased susceptibility to apoptosis.

Clinical Features

Neuromuscular Features

The distinguishing feature of the neuromuscular picture is the early development of flexion contractures of the arm and leg muscles and extension contractures of the cervical muscles. The arms are held in a semiflexed position and cannot be extended (see Figure 35.1 ). The feet are in an equinus position. The neck cannot be flexed onto the sternum, although extension is possible (see Figure 35.2 ). There may also be contractures at the knees, wrists (see Figure 35.3 ), and fingers. In contrast to other neuromuscular diseases in which contractures develop when weakness is advanced, contractures in EDMD develop before any significant weakness occurs, and their severity remains out of proportion to the degree of weakness throughout the course of the disease. Contractures are “in the direction of the weakest joint-moving muscle.” Muscle weakness is mild and slowly progressive. Muscle wasting is most pronounced in the biceps brachii and posterior leg muscles (see Figure 35.4 ). Patients rarely lose ambulation. All forms of EDMD are clinically similar but not identical. The severity of the symptoms can show striking variability even within the same family.

Flexion contracture of arm muscles at the elbow and marked humeral muscle wasting with relative sparing of shoulder muscles in Emery-Dreifuss muscular dystrophy. The arm cannot be fully extended at the elbow and is maintained in a semiflexed position. The pacemaker insertion site is evident (arrow).

Extension contracture of posterior cervical muscles in X-linked Emery-Dreifuss muscular dystrophy. Note the significant limitation of neck flexion (the patient was asked to flex his neck fully onto the sternum).

Extension contracture of forearm muscles at the wrist. The patient (the same one shown in Figure 35.2 ) is unable to flex his wrist beyond a few degrees ( A ), although he can fully extend it ( B ).

Atrophy of posterior calf muscles, particularly of the medial gastrocnemius muscles.

Onset is usually in childhood, when the child is noted to have difficulty with ambulation, usually because of toe-walking due to flexion contractures of the leg muscles. The problem is sometimes noticed as soon as the child starts to walk. Limitation of elbow extension is usually noticed later in the first decade; however, this is the first symptom in some patients. Limitation of neck flexion is usually noted in the second and rarely in the first decade. Muscle weakness starts in the first or the second decade. Mental retardation or even mild cognitive delay is not a feature of the disease.

Examination reveals flexion contractures at the elbows (see Figure 35.1 ), shortening of the Achilles tendons, and extension contracture of the cervical muscles (see Figure 35.2 ). Mild weakness and atrophy of biceps brachii, triceps (see Figure 35.1 ), and posterior calf muscles (see Figure 35.4 ) are noted, although more widespread involvement may be seen. Upper extremity distal muscles can sometimes be weak, and distal contractures may occur (see Figure 35.3 ). Mild facial weakness may be present. Deep tendon reflexes are reduced or absent. Pseudohypertrophy is usually not seen. Scoliosis is rarely observed.

Cardiac Features

Atrial conduction defects are the most common cardiac abnormalities in EDMD 1. First-degree atrioventricular block, sinus bradycardia, and supraventricular tachycardias may be early signs of cardiac involvement. They may progress to atrial flutter, atrial fibrillation (see Figure 35.5 ), and atrial standstill. The onset of cardiac abnormalities usually occurs in the third decade in EDMD. Earlier onset of severe involvement is rare. A recent study reported serious ventricular arrhythmia in a young patient with EDMD 1. Cardiac involvement is more aggressive in EDMD 2; ventricular arrythmias occur earlier and lead to dilated cardiomyopathy. Sudden death can occur, more frequently in EDMD 2, but also in EDMD 1. There is no correlation between the severity of the neuromuscular involvement and the degree of cardiac abnormality.

Electrocardiogram of a 25-year-old patient with X-linked Emery-Dreifuss muscular dystrophy. The main features are a heart rate of 52 beats/minute, irregular rhythm, atrial fibrillation, and incomplete right bundle branch block.

Differential Diagnosis

EDMD first needs to be differentiated from other neuromuscular diseases presenting with contractures in the extremities and rigidity at cervical and dorsolumbar spine. They include congenital muscular dystrophies (CMD), congenital myopathies (such as central core and centronuclear congenital myopathies), and rigid spine syndrome (RSS). None of these disorders shows cardiac conduction defects or blocks.

One might be led to underestimate this differentiation, thinking that EDMD has prominent cardiac conduction defects in contrast to all the other conditions. This is not at all helpful when one encounters a child with a rigid spine because the cardiac defects in EDMD become apparent at a much later age. Thus, differentiation is done on other grounds in early childhood.

There are several forms of CMD, and they present within the first year of life with severe hypotonia accompanied by possible contractures, very high CK, and dystrophic muscle biopsy findings. Symptoms related to the central nervous system might be a part of the clinical picture. Most of them are inherited in an autosomal recessive pattern. Congenital myopathies have distinctive muscle biopsy findings.

The most difficult differential diagnosis is that from RSS, as evidenced by cases previously diagnosed as RSS that have turned out to be EDMD after the advent of molecular biology. Early and severe scoliosis in the first decade is in favor of RSS and should prompt search for selenoprotein 2 gene mutation. Severe respiratory involvement is usually present in RSS but is exceptional in EDMD.

Genotype-Phenotype Correlation

At present, mutations in six genes ( EMD , LMNA , SYNE1 , SYNE2 , FHL1 , and TMEM43 ) are associated with the EDMD phenotype. They cause very similar phenotypes, difficult to distinguish from each other.

The majority of EMD mutations are null mutations with no emerin expression. Missense mutations with reduced expression of emerin are very rare. There is remarkable intra- and interfamilial variation regarding age of onset, severity, and course of neuromuscular and cardiac manifestations, independent of each other. This variation is more pronounced than in other dystrophies. It could be speculated that some modifier gene yet unexplored might have an important role in the modulation of disease severity in nuclear envelopathies.

LMNA mutations are mostly missense mutations. No correlation has been found between the location of the mutation and the clinical phenotype, including severity. In addition to the typical EDMD phenotype, severe phenotypes with very early onset (second year of life) weakness have been reported.

EDMD 2 is phenotypically very similar to EDMD 1, particularly with respect to its neuromuscular features. Onset of the weakness might be earlier and more severe in EDMD 2 and might precede contractures, which are usually the presenting symptoms in EDMD 1. Loss of ambulation might be seen in EDMD 2, which is very unlikely in EDMD 1. Furthermore, patients with EDMD 2 have more severe and progressive weakness of the biceps brachii compared with EDMD 1. Calf hypertrophy might be more common in EDMD 2.

The major distinguishing phenotypical difference of EDMD 2 from EDMD 1 concerns the nature of the cardiac involvement, whereas conduction disturbances, especially atrial arrhythmias, are frequent and the major cardiac problem in EDMD 1, dilation of the left ventricle, cardiac failure, and ventricular tachyarrhythmias appear to be more common in EDMD 2. Cardiac involvement in EDMD 2 is more severe compared to EDMD 1 patients. Isolated cardiac involvement is more frequently found in EDMD 2.

LMNA mutations rarely cause AR phenotypes. In a family with homozygous mutations in the two alleles in the proband and heterozygous mutations in the carriers, the patient had severe neuromuscular symptoms with confinement to a wheelchair, but she had no cardiac involvement even at age 40.

A more severe phenotype of EDMD with unusual disease progression rate is reported in individuals who are carrying mutations in EMD and the desmin gene at the same time.

FHL1 mutations cause a similar phenotype to EDMD 1. FHL1 mutations should be considered in patients with X-linked inheritance pattern, especially if hypertrophic cardiomyopathy is present.

SYNE1 , SYNE2 , and TMEM43 mutations are very rare and it is too early to try to make phenotype-genotype correlations in patients with these mutations.