A multistage sequencing strategy pinpoints novel candidate alleles for Emery-Dreifuss muscular dystrophy and supports gene misregulation as its pathomechanism

Background

As genome-wide approaches prove difficult with genetically heterogeneous orphan diseases, we developed a new approach to identify candidate genes. We applied this to Emery-Dreifuss muscular dystrophy (EDMD), characterised by early onset contractures, slowly progressive muscular wasting, and life-threatening heart conduction disturbances with wide intra- and inter-familial clinical variability. Roughly half of EDMD patients are linked to six genes encoding nuclear envelope proteins, but the disease mechanism remains unclear because the affected proteins function in both cell mechanics and genome regulation.

Methods

A primer library was generated to test for mutations in 301 genes from four categories: (I) all known EDMD-linked genes; (II) genes mutated in related muscular dystrophies; (III) candidates generated by exome sequencing in five families; (IV) functional candidates — other muscle nuclear envelope proteins functioning in mechanical/genome processes affected in EDMD. This was used to sequence 56 unlinked patients with EDMD-like phenotype.

Findings

Twenty-one patients could be clearly assigned: 18 with mutations in genes of similar muscular dystrophies; 3 with previously missed mutations in EDMD-linked genes. The other categories yielded novel candidate genes, most encoding nuclear envelope proteins with functions in gene regulation.

Interpretation

Our multi-pronged approach identified new disease alleles and many new candidate EDMD genes. Their known functions strongly argue the EDMD pathomechanism is from altered gene regulation and mechanotransduction due to connectivity of candidates from the nuclear envelope to the plasma membrane. This approach highlights the value of testing for related diseases using primer libraries and may be applied for other genetically heterogeneous orphan diseases.

Funding

The Wellcome Trust, Muscular Dystrophy UK, Medical Research Council, European Community's Seventh Framework Programme “Integrated European –omics research project for diagnosis and therapy in rare neuromuscular and neurodegenerative diseases (NEUROMICS)”.

Immunohistochemistry on a panel of Emery-Dreifuss muscular dystrophy samples reveals nuclear envelope proteins as inconsistent markers for pathology

•

Altered distribution of EDMD-linked proteins is not a general characteristic of EDMD.

•

Tissue-specific proteins exhibit altered distributions in some EDMD patients.

•

Variation in redistributed proteins in EDMD may underlie its clinical variability.

Reports of aberrant distribution for some nuclear envelope proteins in cells expressing a few Emery–Dreifuss muscular dystrophy mutations raised the possibility that such protein redistribution could underlie pathology and/or be diagnostic. However, this disorder is linked to 8 different genes encoding nuclear envelope proteins, raising the question of whether a particular protein is most relevant. Therefore, myoblast/fibroblast cultures from biopsy and tissue sections from a panel of nine Emery–Dreifuss muscular dystrophy patients (4 male, 5 female) including those carrying emerin and FHL1 (X-linked) and several lamin A (autosomal dominant) mutations were stained for the proteins linked to the disorder. As tissue-specific nuclear envelope proteins have been postulated to mediate the tissue-specific pathologies of different nuclear envelopathies, patient samples were also stained for several muscle-specific nuclear membrane proteins. Although linked proteins nesprin 1 and SUN2 and muscle-specific proteins NET5/Samp1 and Tmem214 yielded aberrant distributions in individual patient cells, none exhibited defects through the larger patient panel. Muscle-specific Tmem38A normally appeared in both the nuclear envelope and sarcoplasmic reticulum, but most patient samples exhibited a moderate redistribution favouring the sarcoplasmic reticulum. The absence of striking uniform defects in nuclear envelope protein distribution indicates that such staining will be unavailing for general diagnostics, though it remains possible that specific mutations exhibiting protein distribution defects might reflect a particular clinical variant. These findings further argue that multiple pathways can lead to the generally similar pathologies of this disorder while at the same time the different cellular phenotypes observed possibly may help explain the considerable clinical variation of EDMD.

Purification of Lamins and Soluble Fragments of NETs

Lamins and associated nuclear envelope transmembrane proteins (NETs) present unique problems for biochemical studies. Lamins form insoluble intermediate filament networks, associate with chromatin, and are also connected via specific NETs to the cytoskeleton, thus further complicating their isolation and purification from mammalian cells. Adding to this complexity, NETs at the inner nuclear membrane function in three distinct environments: (a) their nucleoplasmic domain(s) can bind lamins, chromatin, and transcriptional regulators; (b) they possess one or more integral transmembrane domains; and (c) their lumenal domain(s) function in the unique reducing environment of the nuclear envelope/ER lumen. This chapter describes strategic considerations and protocols to facilitate biochemical studies of lamins and NET proteins in vitro. Studying these proteins in vitro typically involves first expressing specific polypeptide fragments in bacteria and optimizing conditions to purify each fragment. We describe parameters for choosing specific fragments and designing purification strategies and provide detailed purification protocols. Biochemical studies can provide fundamental knowledge including binding strengths and the molecular consequences of disease-causing mutations that will be essential to understand nuclear envelope-genome interactions and nuclear envelope linked disease mechanisms.

LINC'ing form and function at the nuclear envelope

The nuclear envelope is an amazing piece of engineering. On one hand it is built like a mediaeval fortress with filament systems reinforcing its membrane walls and its double membrane structure forming a lumen like a castle moat. On the other hand its structure can adapt while maintaining its integrity like a reed bending in a river. Like a fortress it has guarded drawbridges in the nuclear pore complexes, but also has other mechanical means of communication. All this is enabled largely because of the LINC complex, a multi-protein structure that connects the intermediate filament nucleoskeleton across the lumen of the double membrane nuclear envelope to multiple cytoplasmic filament systems that themselves could act simultaneously both like mediaeval buttresses and like lines on a suspension bridge. Although many details of the greater LINC structure remain to be discerned, a number of recent findings are giving clues as to how its structural organization can yield such striking dynamic yet stable properties. Combining double- and triple-helical coiled-coils, intrinsic disorder and order, tissue-specific components, and intermediate filaments enables these unique properties.