Search for the genetic basis of dyslexia

Dyslexia is a learning impairment that affects between 5-12% of children in the United States. While this disorder is very common, many people have misconceptions about what it actually is. “They think of it as mixing p’s and q’s and d’s and b’s, but all children do that – it’s got nothing to do with dyslexia,” states Dr. Wendy Raskind, professor emerita of medical genetics at the University of Washington. Children with dyslexia have trouble reading and spelling words dictated to them because they cannot connect the sounds they hear with the letters they see or write. So, it’s not that p and q or d and b look similar get mixed up in the dyslexic brain, but rather that the brain cannot connect the sounds of those letters to their written form.

To understand why people with dyslexia have trouble learning to read, it’s useful to have a rudimentary lesson on linguistics. Some languages like English are phonological. Phonologic language speakers learn how to read by learning the sounds that each letter in the alphabet makes. These languages also have distinct phonemes, or unique sounds that are associated with specific letters or combinations of letters. English has 44 distinct phonemes. For example, the s in the word silly and the s in the word treasure make distinct sounds that readers parse apart by putting the s sound in the context of all the other sounds in the word. To complicate things, some phonemes are two-letter combinations (for example, how the phoneme th is different from the phonemes t and h).

Importantly, dyslexia does not affect comprehension. “If you read them a story and ask a question, they’ll understand and answer the question. They just can’t read the story!” explains Raskind. While the symptoms of dyslexia can be alleviated through educational intervention, other aspects of the condition like slow reading speed and poor spelling can persist into adulthood.

Scientists estimate that about 50-75% of the risk for dyslexia can be attributed to genetic factors. Generally speaking, there is no single gene alteration that causes dyslexia. Rather, it is influenced by variation in multiple genes. Over the years, many groups have discovered suspected dyslexia-linked genes by a variety of genetic approaches, including association studies, linkage analyses, and studies of genes disrupted by chromosome translocations. Differences in the study population selection, the ways researchers defined dyslexia, and the dyslexia-related traits were evaluated may explain why different research groups have had trouble confirming other groups’ proposed candidate genes. Additionally, most researchers only studied the protein-coding genetic regions that they know how to interpret. This means that information about regulatory regions of DNA that might control the protein abundance or expression have remained unexplored. Raskind and her team thought that previous approaches had not been deep enough to rule out involvement of the candidate genes or to identify the changes in those genes that might be the culprits.

To this end, the team took advantage of a collection of DNA samples and phenotyping data from over 2000 participants in families who had at least one child with a diagnosis of dyslexia from the University of Washington, the Hospital for Sick Children in Toronto, and the University of Houston. They sequenced the protein-coding region and regulatory regions of five genes already proposed as candidates for dyslexia risk. This experimental design overcame some weaknesses in previous studies. For one, enrolling participants from separate institutions with subtle differences in ascertainment, testing protocols and diagnostic criteria results makes it more likely that the results will be generalizable. Second, analysis of quantitative dyslexia-associated traits, like scores on a test of reading isolated non-real words, rather than a categorical diagnosis of dyslexia bypasses complications of differences in criteria used to make such a diagnosis. Third, they took advantage of a high-throughput sequencing approach to do a more comprehensive analysis of the genes that included some regulatory regions as well as the protein coding regions. Regulatory DNA sequences can affect the timeframe, abundance, and location of each gene’s expression, crucial data that was missing from previous analyses. They realized that even with the large number of participants, to ensure the ability to obtain significant results given the large amount of DNA being analyzed, they would have to limit the number of genes studied to five.

So, what did they find? For the gene DNAAF4, they found that there was no significant correlation with any of the six dyslexia traits they examined. This was somewhat surprising, as several previous studies have linked this gene region to dyslexia. “We can’t say it’s not involved in any way…maybe there’s something in a different region of the regulome, but we don’t find anything in the coding region,” says Raskind. The team next looked at CYP19A1, another gene on the same chromosome as DNAAF4 that encodes an enzyme that converts testosterone to estrogen and is active in the brain. This gene is an interesting candidate because of more males than females are diagnosed with dyslexia. A common variant in that gene was associated with the ability to sound out non-words, real words, and spelling performance.

The group next examined two closely linked genes on a different chromosome. They found that one variant of the DCDC2 and KIAA0319 region was associated with reading speed. The team was unable to distinguish any associations for either individual gene because of the complexity of the region, but this work and other published work about these genes in the brain suggests a potential role for the pair in dyslexia. Finally, they examined the gene GRIN2B. Severe alterations in this gene are associated with neurodevelopmental problems, so the team had reason to suspect that minor alterations might contribute to dyslexia. Indeed, they found that rare GRIN2B variants affected spelling test performance in their participants.

Overall, Raskind and her team believe that this work provides a strong argument to consider the coding and regulatory regions of a gene to fully understand its effect especially in complex disorders that do not affect overall health. These associations provide rationale to examine the roles of these genes in more controlled settings like cell culture or animal models. “Whatever the root causes of dyslexia are, they will overlap with other learning disabilities and potentially other, more severe problems, and if we can get a toehold into that, then we get a biochemical pathway,” said Raskind of the future applications for this study. Raskind hopes that future advances in understanding gene regulation will provide additional insights into fully decoding the genetic basis for dyslexia and conditions like it.

This work was supported by the Eunice Kennedy Shriver National Institute of Child Health and Development, the Canadian Institutes of Health Research, and the Hospital for Sick Children Research Training Program.

Fred Hutch/UW/Seattle Children’s Cancer Consortium member Dr. Wendy Raskind contributed to this work.

Chapman NH, Navas PA, Dorschner MO, Mehaffey M, Wigg KG, Price KM, Naumova OY, Kerr EN, Guger SL, Lovett MW, Grigorenko EL, Berninger V, Barr CL, Wijsman EM, Raskind WH. 2025. Targeted analysis of dyslexia-associated regions on chromosomes 6, 12 and 15 in large multigenerational cohorts. PLoS One. 20(5):e0324006. doi: 10.1371/journal.pone.0324006.