Summary
NMD-CHIP addresses two significant issues facing neuromuscular genetics: diagnosis of mutations in genes already known to be involved in NMDs; and discovery of new genes not previously known to be responsible for NMDs.
Obtaining a genetic diagnosis is often a lengthy and expensive process involving testing of many genes one by one until the mutation is found. Chip technology can allow all the potentially relevant genes in a patient’s DNA to be tested simultaneously, which is a much faster and cheaper process.

The project is designing chips for the diagnosis of mutations already known to cause Duchenne / Becker muscular dystrophies (DMD/BMD), limb girdle muscular dystrophies (LGMD), congenital muscular dystrophies (CMD), and hereditary motor-sensory neuropathies or Charcot-Marie-Tooth neuropathies (CMT). These diagnostic chips are described as "known-gene chips".

However, another problem facing doctors and patients is that not all genes responsible for neuromuscular disorders have yet been discovered, and an estimated 30 to 40% of patients are left without a diagnosis even once all known genes have been tested. The NMD-chip project is also developing "candidate-gene chips" to look for mutations in patients who have a neuromuscular disorder where the gene responsible has not yet been discovered.

Strategy
The scientific strategy of the project is to design 4 types of chip: two for known genes, and two for candidate genes. In each case, both comparative genomic hybridization (CGH) arrays and sequence capture chips are being developed – the former to test for insertions and duplications and the latter for point mutations.

The technology used, developed by Roche-Nimblegen, allows up to 2.1 million probes to be spotted on one chip, and the resulting images to be analyzed with a 2-micron resolution scanner.

The project is making use of the new generation of 12-plex chips, each chamber comprising 72,000 probes covering all the exons, intron flanking regions, and where known, deep intronic mutation spots of genes for each group of pathologies (CMD / LGMD / DMD / CMT).

Project Management
The NMD-Chip Kick-Off Meeting was help in Paris on November 18-19th 2008. Since then, the steering committee has met approximately every 6 months as follows:

Marseille, France – 6-8th April 2009
Stockholm, Sweden – 31st August – 2nd September 2009
Ferrara, Italy – 29-31st March 2010
Wuerzburg,Germany – 4-6 October 2010

At these meetings partners were able to collaborate, share information discuss progress and make decisions about the project’s progress together.

Chip design – comparative genomic hybridization (CGH) arrays
Sequences for the 50 genes known to be implicated in LGMD, congenital muscular dystrophies and congenital myopathies and the 43 genes implied in the CMT group of diseases were selected.

The information about these genes and sequences was included in a global database together with other information collected from the partners to create a database for chip design. This was released in May 2009.

First-generation CGH chips for known genes have been validated. Two CGH-arrays containing probes covering all genes known to be involved in NMDs have been developed. The first is dedicated to muscular dystrophies (including both dominant and recessive LGMDs), congenital muscular dystrophies, and congenital myopathies. The second is dedicated to hereditary motor sensory neuropathies. These chips will now become part of the standard diagnostics workflow to progressively replace the current techniques, although they may still be improved through re-design.

Chip design – sequence capture (SC) chips
Lessons learned from CGH-array design were used to improve the design of SC chips which was initiated after validation of the CGH arrays. The first sets of SC chips are being validated.

Results for LGMD/DMD/CMD sequence selection will be compared to another capture technique "in solution" called SureSelect from Agilent. It has been suggested that this may give better results. Both technologies will be evaluated, and the better one will be selected for the next steps of the project.

Chip validation (known-gene chips)
The validation of the NMD-CHIP 12-plex array has been performed by several partners.

Nicolas Levy’s team in Marseille have validated the LGMD/CMD chip using the  following controls:

• DYSF and CAPN3 mutations
• 3 sizes: DYSF homozygous deletion ; DYSF heterozygous deletion; CAPN3 heterozygous deletion

Their CMT chip validation used the following controls:

• One female affected with CMT1A, with “classical” 1.5 Mb CMT1A duplication
• One male with 1.5 Mb duplication affected with HNPP
• One female with partial deletion affected with HNPP
• One female affected with CMT1A with partial duplication
• Patients were diagnosed by use of STRs, MLPA, Q-PCR
• DNA from a non CMT patient was used as control "negative" DNA

Jamel Chelly’s team at the Paris-Cochin Institute has performed several tests on the LGMD/CMD chips with negative and positive control samples:
A group of 39 patients or female carriers with known exonic deletions or duplications were tested by array CGH. The rearrangements were previously identified by other techniques:

• 26 DMD/BMD 
• 3 gamma-sarcoglycanopathies (SGCG)
• 1 alpha-sarcoglycanopathy (SGCA)
• 1 calpaïnopathy
• 2 dysferlinopathies
• 4 emery-Dreifuss (EMD) 
• 2 FHL1 mutated patients

Christophe Beroud’s team in Montpellier have worked on patients with recently identified large rearrangements in the LAMA2 gene responsible for DMC1A (unpublished confidential data). These new findings gave them the opportunity to validate the array on a wider set of genes with known mutations.
The validation process has been conducted in parallel with another Roche 12-plex custom array that has been specifically designed for Montpellier’s diagnostic laboratory.

The purposes of the validation process were to evaluate the potential of the NMD-CHIP 12-plex array to:

a) efficiently detect rearrangements of the LAMA2 gene;
b) efficiently define the breakpoints of these large rearrangements and
c) to compare its performances to another Roche custom 12-plex array with different gene coverage.

All results have been validated using other techniques such as quantitative dHPLC and semi-quantitative multiplex PCR.

At the NIEH laboratory (Veronika Karcagi) similar work has been done. Here they have compared the 12-plex NMD-CGH microarray containing probes for known genes of LGMD, CMD, DMD diseases with the results of the whole genome 12-plex microarray from Roche. NIEH performed studies with isolated DNA samples of NMD patients with known exonic deletions or duplications. All the experiments were performed in parallel and with similar conditions.

INSTYMOL (Gisele Bonne) have started validating the 12-plex array using control DNA of patients with known NMDs. In particular, 22 DNAs have been tested on 4 chips including: 9 DMD, 3 EMD, 1 FHL1, 2 LARGE, 1 POMT2, 1 POMGnT1, 1 RYR1, 2 SGCG, 1 COL6A1, 1 COL6A2.

UCL (Francesco Muntoni) is currently carrying out validation steps of the 12-plex array using positive control DNA from patients with known NMD and their own DNA samples. They have 26 samples divided into three subgroups:

• Patients with LAMA2, COL6A2 and POMT1 heterozygous and homozygous deletions involving whole exons; 
• Patients with mutations in the POMT1, LARGE, SEPN1, COL6 and RYR1 genes – this is in order to determine whether the NMD-chip can detect mutations that involve the insertion or deletion of <30bp;
• Patients with whole exon deletions and duplications in the DMD gene that have previously been analysed using a custom-made DMD-array – this is in order to compare quality of the data.

At the University of Ferrara (Alessandra Ferlini) the validation of the specific NMD-Chip using control DNA of patients with known dystrophin mutations is currently in progress.

Candidate-gene chips validation
The candidate-gene chips have been designed and are currently under validation. These chips aim to identify new genes involved in the three groups of diseases of the project, namely congenital muscular dystrophies (CMD), Charcot-Marie-Tooth diseases (CMT) and limb-girdle muscular dystrophies (LGMD).

309 candidate genes for CMT were selected by the expert scientists using data from the literature and bioinformatics tools. The genes were chosen for their implication in well known pathways, in the physiology of the Schwann cell or the axon, or for their inclusion in protein families or interactions with proteins already known to be involved in CMT. This was done with the input of partners from INSERM.

The CMD list has 345 functional candidate genes mainly selected by using the web based tool Endeavour : http://homes.esat.kuleuven.be/~bioiuser/endeavour/endeavour.php; (Aerts et al., 2006).

Using experimental data, data from the literature, bioinformatics tools and consortium members’ expertise, an extended list was built of 1655 LGMD candidate genes, and a reduced list of 467 LGMD candidates; 245 with the highest priority and 26 suggested MIRgenes. This was done with the input of partners from GENETHON, NIEH, InstMyol.

Bioinformatics
Several software tools have been created to improve the chip design, to collect data, and to analyse the mutations detected by the chips. Partners have been adding data to the database and by March 2010 it contained more than 115 data sets. The entries are checked to ensure they have been added accurately.

Website
A project website has been set up at www.nmd-chip.eu which enables dissemination of information about the project to stakeholders.

Societal aspects
A project ethics council has been established in order to advise and answer questions from NMD-CHIP partners and stakeholders. The PEC details are listed on the website along with links to relevant documentation (such as the Human Tissue Act) and other organisations dealing with bio-ethics.