Gene deletion versus mutation (variant): sometimes missing a gene is better.

David in refrig age 4.5
David (age 4 1/2) teaching himself to
stand up using the refrigerator shelf

Originally posted 1 July 2015
Updated 21 July 2021
Available in Portuguese

Cerebellum and walking

It took David about 6 years to learn how to stand and walk. Even then, he was very unstable. I remember, if he was standing in a room and the lights went out, he would simply fall over. His balance was totally dependent on vision. He had not yet learned to use inner ears and ankle joint position sense to maintain balance. That is what the cerebellum does. It brings together the three senses involved in balance and makes rapid adjustments to our posture. Not many animals are upright walkers and most of the other ones are birds. They have wings! There is no doubt that David has problems with proper functioning of his cerebellum (including seizures associated with the cerebellum), but he did, eventually, learn to walk.

Cerebellar ataxia 10 (SCA10) and the ATXN10 gene

Ataxia is poor coordination, usually associated with walking. Drunken sailors (and others) walk with a staggered gait because alcohol affects the cerebellum. There are many different genes that can severely affect cerebellum function and thus walking. Variants of these genes produce a group of syndromes called “spinocerebellar ataxia” or SCA. Different genes cause different SCAs, each with its own characteristic phenotype.  Mutations (properly called “pathogenic variants”) of the ATXN10 gene lead to SCA10 (Zu et al., 1999). These patients have ataxia (poor gait), dysarthria (poor speech) and nystagmus (poor eye control) — all cerebellar functions. Many individuals with SCA10 also have seizures. The condition is inherited and found primarily in Latin American families (Teive and Ashizawa, 2014). The disease has a late onset (age 25 to 45) and starts out slowly. Over time, it can be totally incapacitating.

This blog is about 22q13 deletion syndrome (Phelan-McDermid syndrome, PMS). Half of all individuals with 22q13 deletion syndrome have deletions larger than 5 Mb (see my posting Understanding deletion size), and they are missing the ATXN10 gene. If I did not know more about SCA10, I would be frightened for David. He is currently between the ages of 25 and 45 and SCA10 is a frightening disease. However, it has been shown that SCA10 is not caused by missing the ATXN10 gene. Only specific variants of ATXN10 cause SCA10 (Karen et al., 2010). These rare variants of the ATXN10 gene create a protein that has gained a new function (McFarland and Ashizawa, 2012). This process is called a “gain-of-function mutation”. Don’t be fooled by the terminology. In this case, gain-of-function is a bad thing. The new function kills Purkinje neurons in the cerebellum (Xia et al., 2013).

Better off

About 50% of people with 22q13 deletion syndrome are missing the ATXN10 gene. The impact of missing this gene has never been studied on our children, but right now it looks like they are far better off missing the gene than having the rare variant found in families with SCA10.

One interesting, related observation comes from the SHANK3 gene. There is a growing body of evidence that some SHANK3 variants may be gain-of-function. Mouse models with SHANK3 mutations suggest that different mutations do very different things. (Note, the mice are genetically engineered, so it is proper to call the modified gene a mutation.)  Similar to these mouse models, in humans, different variants of the SHANK3 gene have different effects. In many cases, rare SHANK3 variants have no effect. There are people in the general population with SHANK3 variants who will never suspect their gene is unusual. Their variants are not pathogenic. Other SHANK3 variants cause intellectual disability, autism spectrum disorder or both.  It would be helpful to study human cases where all of SHANK3 is missing, but no other gene. That is the best way to examine the precise impact of missing SHANK3. Once we understand the precise impact of SHANK3, we can unravel the complex effects of all the genes of Phelan-McDermid syndrome. Indeed, that is perhaps the most important goal of PMS research.

If we want to understand Phelan-McDermid syndrome we need to understand SHANK3. Unfortunately, we are not making good use of our available data. Many of the studies I have read indiscriminately combine SHANK3 variant cases with SHANK3 complete deletion cases. That approach ignores the realities of rare variants and partial gene deletions, as explained above. I have also read papers where the authors indiscriminately combine deletions of all sizes to infer the impact of a SHANK3 deletion, without accounting for the impacts of the many other genes that contribute to Phelan-McDermid syndrome. I have written several blogs on why this approach is unsound (see PMS, IQ and why interstitial deletions matter and Which PMS genes are most important? ). What we do know is some people with Phelan-McDermid syndrome who are missing SHANK3 are more functional than some people with rare pathogenic SHANK3 variants. That strongly suggests that in some cases a person is better off missing the entire SHANK3 gene rather than having a specific pathogenic variant. Sometimes missing a gene is better.


Some previous posts:
Is 22q13 deletion syndrome a ciliopathy?
Understanding translocations in 22q13 deletion syndrome: genetics and evolution
Understanding deletion size
22q13 deletion syndrome – an introduction