‘Big data’ projects require detailed analysis of unimaginably large volumes of complex data. This is especially true in the realm of MND gene discovery when searching for MND-associated genes – where the greater the number of samples analysed, the greater the possibility of finding the relatively less frequently occurring genetic causes (known as ‘rare’ variants). Literally a needle in a haystack.
These discoveries are no less important, as each new discovery is highly significant and provides another piece of the puzzle in our understanding of the causes and avenues to target for potential treatments. A clear example of this is collecting and mining the data from tens of thousands of human ‘genomes’ to identify the genes responsible for MND. By working together, researchers can greatly increase their ability to tease out the difficult to find discoveries.
A specific example of this extraordinary teamwork and global collaborative approach is the identification of the most recently found MND/ALS gene. This recent finding, published last month in the journal Neuron (link), is the result of a very large collaboration of over 200 authors from MND research institutions from 5 different countries and 9 consortia across Europe, Australia and the US, that collaborated openly to combine their analytical skills and MND patient genome sequencing data to identify this MND/ALS gene. The consortia included the MND Association-supported Project MinE , an international collaboration aiming to analyse over 20,000 DNA samples.
Each MND causative gene that is identified provides further understanding of the pathways and underlying causes of motor neurone degeneration. Understanding how these mechanisms work (or fail in people with MND) will ultimately lead us to potential treatments and a world free from MND.
The gene identified in this paper is KIF5A (which stands for Kinesin family member 5A). The discovery of KIF5A provides an additional specific target and pathway (cytoskeleton and axonal transport) to focus on the search for potential treatment and drug development. KIF5A is a cargo transporter that is needed for axonal transport.
Several neurodevelopmental and neurodegenerative diseases occur as a result of mutations in proteins of the axonal transport machinery. Axonal transport has to continue to run efficiently throughout our lives to maintain the health and function of the neurone, which is subject to all kinds of ‘attack’ from stress and environmental impacts.
Efficient axonal transport is essential to the function and health of motor neurones which, unlike many cells, cannot regenerate. Axonal transport is the process that moves materials that are essential to keep the cell working. These materials are moved from the cell body of the neuron to the axonal terminal of neurons, along the cytoplasm of its axon.
This is quite a journey, since some motor neurones can be up to 1 metre long (eg a motor neurone running from the spine to the foot). To put this in perspective, the cell body of this motor neuron would be approximately 100 microns (0.1 millimeter) in diameter, whereas the axon is about 1 meter (1,000 millimeters) in length. So, the axon of a motor neuron is 10,000 times as long as the cell body is wide. If you visualise the cell body of the nerve cell to be the size of a ping-pong ball, your axon would be 400 meters long (the full distance of an athletics track, or three football pitches!).
KIF5A is a kinesin – a protein required for transport
Kinesins are a key player in axonal transport; they transport vital cellular cargo, such as important proteins, along the length of neurones. Kinesins are like ‘trucks’ carrying fuel and supplies along roads from the suppliers to a remote factory. The efficient functioning of the trucks is needed for the transport of the fuel and supplies to happen – to run the factory.
If the trucks are damaged and can’t function properly then everything at the factory stops, and eventually closes down – as does the neuron. KIF5A is known to transport important proteins, some of which may be MND causative if not correctly transported. The authors speculate that mutations in the kinesin KIF5A protein (the ‘truck’) causes disease by disrupting the normal axonal transport process.
Rare genetic variants can only be detected in large numbers of samples, and therefore pooling data from many scientific groups is required for rare variants to be discovered.
These gene discoveries are no less important than more commonly occurring mutations such as SOD1 (found in 10% of familial cases), as each and every piece of the puzzle is vital in putting together the true picture of the mechanisms of MND and neurodegeneration.
How was KIF5A identified?
Two approaches were used and each pointed to KIF5A as a gene associated with MND. One analysed common variants while the other examined rare variants. The first was a Genome-Wide Association Study (GWAS). GWAS typically uses SNP chip data and compares (relatively) common variants between case and control cohorts. In this case, approximately 80,000 DNA samples, including 25,000 people with MND were analysed. The second approach was ‘Rare Variant Burden Analysis’.
Rare variants are mutations that may only occur and be responsible for a relatively low number of cases. Rare variants may account for the as-yet unidentified MND genes (missing heritability) seen in ALS (approximately 30% of familial MND cases have no identified genetic cause (yet)). GWAS can miss rare variants. KIF5A was identified but close to the cut-off threshold. Therefore the second approach used was ‘Rare Variant Burden Analysis’. Rare variant studies use whole genome (WGS) or exome sequence data. In this case exome sequencing of approximately 19000 controls and 1000 people with MND which allowed the identification of this more difficult gene to associate with MND because it may only be responsible for a relatively low number of cases in the population.
One way of explaining GWAS vs rare variant analysis is like going on to Google Earth – the GWAS is a low-resolution picture of a wide area (say, a town) that helps spot the key features such as buildings, cars in carparks etc., whereas rare variant analysis ‘zooms in’ to give a much higher level of detail (say, showing which model of cars are in the car park, to occasionally identify a less common vehicle, such as a classic car).
In summary, the study showed that:
- Mutations at the cargo-binding ‘tail’ end (known as the C-terminus) of the KIF5A protein cause MND.
- 12 different MND-causing mutations in the tail end of the gene were identified.
- Mutations in the gene that result in a non-functional KIF5A protein cause MND but these people with MND have a relatively early onset and longer survival (onset 45 years, survival 10 years) compared to typical MND prognosis (average onset 65 years of age and survival 3 years).
