Research Projects

The genetics of SLI

Specific Language Impairment (SLI) is a complex genetic disorder. Complex disorders are caused by interactions between many genes, spread across different chromosomes.

The genes of each individual are slightly different and it is this variation that determines what you will look like and, to a certain extent, your chances of developing common illnesses such as heart disease. We think that some of these normal gene variations cause people to be more prone to language problems: the more of these susceptibility variants you carry, the more likely you are to develop SLI. This means that the exact combination of genes that are causing SLI may vary between different people. It also means that, in some cases, people who have a copy of some SLI susceptibility genes will still not develop SLI. This makes the detection of genes more difficult than it is for a simple (single-gene) disorder.sequence

Over the past 15 years or so, there has been remarkable success tracking down genes for single-gene disorders such as cystic fibrosis, Duchenne muscular dystrophy and Huntington's disease. But the approaches used for single-gene disorders are not usually powerful enough for complex genetic disorders, such as SLI or autism. Recently, however, novel strategies have been developed that are more appropriate for finding genes influencing complex disorders. We have used these techniques in our research, with the aim to identify the specific genes implicated in SLI.

In the Newbury group, we research the genetic basis of Specific Language Impairment (SLI). This is achieved by three integrated projects which are described below:

   (i) Focussed investigations of chromosome 16

   (ii) The Robinson Crusoe project

   (iii) Genome-wide association and sequencing project

Background studies

IBDWe know that Specific Language Impairment (SLI) runs in families and that this is partly due to genes passed on from parents to children. But as we do not know what causes SLI, we do not know what kinds of genes we should be searching for. In the first stage of our study, we therefore started by trying to identify which chromosomes the genes might lie on. We did this by carrying out a complete genome scan of all 23 human chromosomes.

As part of the SLI Consortium (SLIC) we collected DNA from 98 families with family members affected by SLI. We then sampled the DNA at random intervals to look for genetic regions that showed increased similarity in affected sibling pairs. This is known as a linkage technique and is based on the premise that, if affected children share the genes that cause SLI, then at the position of these genes, they will be genetically more similar than we would normally expect unrelated people to be.

We identified two regions that we think might contain genes that make people prone to SLI: SLI1 (OMIM 606711) on chromosomes 16 and SLI2 (OMIM 606712) on chromosome 19. The results of the first stage of our project were published in 2002 [pdf].

However, a genome screen is simply a statistical method. This means that, at best, it will only enable us to say that a certain genetic region is likely to contain a gene involved in the disorder under study. The more families we are able to include in the study, the more confident we will be that we are focussing on the correct genetic regions.


GWAarrayIn the second stage of our study we therefore included an additional 86 families with family members affected by SLI. Once again we found that regions of chromosomes 16 and 19 were more similar in individuals with language problems than we would expect by chance alone, therefore making it very likely that these chromosomes are really of general importance in SLI. The results of the second stage of our project were published in 2004 [pdf]. In two further studies of additional sample sets, we found more evidence to support the involvement of these two genetic regions in susceptibility to language impairments [pdf1, pdf2].

The genome screen and subsequent replications allowed us to narrow down the search for SLI susceptibility genes from 23 chromosomes to 2 regions on 2 chromosomes: SLI1 (chromosome 16) and SLI2 (chromosome 19). We then began a detailed study of the SLI1 region on chromosome 16. We looked at thousands of common genetic variants across this region and found clusters of variants in two genes (known as ATP2C2 and CMIP) that we think contribute to SLI susceptibility.


We think that differences in the genetic sequence around ATP2C2 and CMIP may alter the way in which these genes work and, in some people, this may decrease their memory span making it more difficult for them to learn new language. We found that the decrease in memory span is particularly strong in individuals with language impairment, and we believe that this is because these individuals also carry additional genetic risk factors that interfere with other biological processes important for language development. It is likely that combinations of risk factors across multiple developmental pathways (e.g. memory) make some individuals particularly prone to language impairments. This research was published in the American Journal of Human Genetics in 2009 [pdf & supplementary data].

Current Research

Our current research is focussed upon the genetic sequence around ATP2C2 and CMIP. We are looking in more detail at the genetic variation in and around these genes to try to find how they can alter the way in which the genes work. Once we understand this, we hope to perform further research to help us to understand how these genetic variants alter the function of particular protein networks. This research will provide important information about the causes of SLI and will provide a starting point in the identification of the biological processes involved in both SLI and normal language development. We hope that an understanding of these systems will help the early identification of affected individuals and allow the design of better speech therapy programs.

The chromosome 16 project is funded by the Medical Research Council (MRC).

The Robinson Crusoe ProjectFIGURA 1 mapa isla en colores

The Robinson Crusoe Island is a small Island approximately 700km off the West coast of Chile. It is geographically and socially isolated and has approximately 600 inhabitants. Studies by researchers at the Univeristy of Chile have found that the inhabitants of this island have a particularly high rate of language impairments. 35% of the children studied had SLI and a further 27.5% had language impirments that were secondary to other developmental conditions. Because of it's relative isolation, the DNA of island inhabitants shows less variability than we would expect in other populations. This can give us extra power to identify causal genetic variants.

Our collaborators have collected DNA from the residents of the Island and we have used these samples to peform another genome screen. This screen highlighted several genetic regions that were more similar in affected individuals than we would expect by chance alone [pdf]. These included regions of chromosome 7 and 13, both of which have been shown to be important in language impaired samples by other groups.

The Robinson Crusoe project was funded by the John Fell Fund and is now funded by the Medical Research Council.

Additional Research

In addition to our investigation of chromosome 16, we have recently identified a third candidate gene on chromosome 7 known as CNTNAP2. This gene is regulated by the FOXP2 protein and so demonstrates a direct link between Dr Simon Fisher's research of verbal dyspraxia and our studies of SLI [pdf & supplementary data].

We are also working with Dr Fisher's lab at the Max Planck Institute for psycholinguistics to perform more in-depth genome screen studies of DNA samples collected from individuals with SLI. We hope that this will help to characterise the SLI2 region of chromosome 19 and perhaps to identify new genetic variants that contribute to SLI susceptibility.