Boys with autism who carry rare, spontaneous mutations have lower intelligence quotients (IQs) and more severe symptoms than do those who may have inherited the disorder. The finding, published 21 October in the Proceedings of the National Academy of Sciences1, hints at two classes of autism risk with varying severity.
“Understanding how the genetic heterogeneity is actually leading to phenotypic heterogeneity is a very exciting time in our research,” says lead researcher Mark Daly, associate professor of medicine at Harvard University.
The researchers looked at harmful mutations present in 618 boys with autism but not in their parents or unaffected siblings.
The researchers initially included 152 girls with autism in their study. This number turned out to be too low to spot any meaningful trends between spontaneous mutations and IQ.
However, these girls have double the rate of harmful mutations compared with the boys. This is in line with the finding that girls need a bigger genetic hit to develop autism than boys do.
Boys with IQs below 100, the population average, have a high rate of spontaneous, or de novo, mutations, the study found. They also have more severe symptoms than boys with higher IQs. This suggests that harmful mutations track with both IQ and autism severity.
By contrast, boys with autism who have IQs higher than 100 have the same number of de novo mutations as do people without autism. This suggests that themutations do not contribute to autism in these boys.
The boys with higher IQs also tend to have a family history of psychiatric disorders, suggesting that inherited milder mutations may have combined to cause their autism.
“It seems as if the higher-IQ boys [have] a different class of genetic mechanism,” says Michael Wigler, professor at Cold Spring Harbor Laboratory in New York. Wigler was not involved in this study, but has found similar results.
In a study published last week, he and his colleagues reported that de novo mutations are present predominantly in children with low IQs. One harmful de novo mutation in an autism gene can lower IQ by about 5 points, and two mutations by about 20 points, they found.
Wigler’s study also found that de novo mutations may account for just under one-third of the risk for autism. This is the latest in a string of results suggesting an important role for de novo mutations in autism risk.
Much of the search for harmful de novo mutations so far has focused on families that have only one member with autism. For example, the new study and Wiger’s report last week both drew samples from the Simons Simplex Collection (SSC) — a database of children with autism whose parents and siblings do not have the disorder2. (The SSC is funded by the Simons Foundation, SFARI.org’s parent organization.)
By comparing the exomes — the protein-coding regions of the genome — of these family members, researchers have tried to pinpoint de novo mutations in the child with autism.
Looking instead at boys with autism who have low IQs may be “markedly more productive,” says Elise Robinson, instructor in medicine at Harvard University and one of the researchers on the new study.
Boys who were unable to complete an IQ test also have an elevated rate of de novo mutations, the study found. These boys have the most severe behavioral problems, further reinforcing the link between de novo mutations and autism severity.
Not everyone with a de novo mutation in an autism gene has severe symptoms, however. And many people with the disorder may carry both de novo and inherited risk factors.
“We shouldn’t set up a situation where we say only the kids who have severe de novo mutations are the ones who are going to be severely affected,” says Brian O’Roak, assistant professor of molecular and medical genetics at Oregon Health and Science University, who was not involved in the new study. “You can have the same high-risk genetic mutation in kids who have an IQ above 100 and on the other side have a kid who is basically on the low IQ end.”
As a proxy for inherited mutations, the researchers looked at the children’s family history — up to first cousins — of depression, bipolar disorder and schizophrenia. This is “clever,” says Kathryn Roeder, professor of statistics at Carnegie Mellon University in Pittsburgh, who was not involved in the study. “It’s a good insight. You see that the people with a family history [of psychiatric disorders] still have autism, but they are high functioning.”
The milder symptoms, which may be the result of common variants found throughout the population, may shed light on important aspects of autism, she says.
“Studying the impact of the common variants will give us greater insight into how we develop into social beings,” Roeder says. “They’re going to get us to the subtle symptoms, which are the pretty interesting part of autism.”
1: Robinson E.B. et al. Proc. Natl. Acad. Sci. USA 111, 15161-15165 (2014) PubMed
2: Iossifov I. et al. Nature Epub ahead of print (2014) Abstract