News The latest developments in autism research.

Girls protected from autism, study suggests

by  /  10 March 2014
Julia Yellow


It takes more mutations to trigger autism in women than in men, which may explain why men are four times more likely to have the disorder, according to a study published 26 February in the American Journal of Human Genetics1.

The study found that women with autism or developmental delay tend to have more large disruptions in their genomes than do men with the disorder. Inherited mutations are also more likely to be passed down from unaffected mothers than from fathers.

Together, the results suggest that women are resistant to mutations that contribute to autism.

“This strongly argues that females are protected from autism and developmental delay and require more mutational load, or more mutational hits that are severe, in order to push them over the threshold,” says lead researcher Evan Eichler, professor of genome sciences at the University of Washington in Seattle. “Males on the other hand are kind of the canary in the mineshaft, so to speak, and they are much less robust.”

The findings bolster those from previous studies, but don’t explain what confers protection against autism in women. The fact that autism is difficult to diagnose in girls may mean that studies enroll only those girls who are severely affected and who may therefore have the most mutations, researchers note.

“The authors are geneticists, and the genetics is terrific,” says David Skuse, professor of behavioral and brain sciences at University College London, who was not involved in the study. “But the questions about ascertainment are not addressed adequately.”

Genetic burden:

The new study draws from the Simons Simplex Collection (SSC), a database of families that have one child with autism and unaffected parents and siblings. (This project is funded by the Simons Foundation,’s parent organization.) In a 2011 study, researchers found that girls with autism in the SSC tend to have more large duplications or deletions of regions of the genome, called copy number variants (CNVs), than do boys with the disorder, although this disparity does not reach statistical significance2.

For the new study, Eichler and his colleagues cataloged the number of CNVs in 109 girls and 653 boys with autism from the SSC. They found that females are twice as likely as males to carry CNVs that are at least 400 kilobases long. (The larger the CNV, the more likely it is to disrupt important genes.)

When the researchers analyzed only CNVs that encompass risk genes for neurodevelopmental disorders, they found that females with autism are three times as likely as males with the disorder to carry CNVs that encompass these genes.

Females with autism also carry slightly more rare mutations that change a single DNA nucleotide than the men do. These are the “nastiest of nasty mutations,” says Eichler, because they interfere with the protein’s function.

The researchers saw a similar but smaller effect for CNVs in a larger group of 9,206 males and 6,379 females referred for genetic testing: 75 percent of this group turned out to have developmental delay, intellectual disability or autism.

Women in this group are 1.28 times more likely than men to carry large CNVs that include risk factors for these disorders.

Many autism-linked mutations arise spontaneously, or de novo, and about 80 percent of these come from the father.

Eichler and his colleagues found that women are far more likely than men to transmit the inherited mutations that confer autism risk.

Of the 27 large CNVs the researchers identified in the SSC group, 70 percent, or 19,were inherited from the mother. Mothers had similarly passed down about 57 percent of the 3,561 CNVs detected in the neurodevelopmental group.

Eichler intends to extend this work in a bigger study to assess whether certain mutations are more likely than others to be inherited.

“I think it’s really critical to identify these inherited components,” he says. “We know they’re there, but we need to really focus on identifying the specific genes so we can advise [parents] a little more about recurrence.”

However, it’s unclear whether this gender bias is the result of genetics or reflects differences in diagnosis or the way females manifest symptoms of the disorder. Girls with autism tend to actively compensate for their symptoms in ways that boys don’t, which may account for the discrepancy, says Skuse.

As a result, the females enrolled in studies may tend to be severely affected and carry multiple mutations. “There is some suggestion that higher-functioning females are out there in the general population, but they’re not being referred,” he says.

The study also does not address why women with autism transmit more mutations, or how they are protected from autism.

“We need to ask what it is about brain development that makes it such that females are protected — because ultimately that is what we want know,” says Aravinda Chakravarti, director of theCenter for Complex Disease Genomics at the Johns Hopkins University School of Medicine in Baltimore. “We need to re-create that developmental environment.”

The most obvious explanation for autism’s gender bias is that because men have only one X chromosome, they are hypersensitive to mutations in this chromosome. In line with this theory, several autism-linked genes are located on the X chromosome. However, most of the mutations that show a gender bias in the new study are not on the X chromosome, suggesting that other factors must be involved.

Correction: This article has been modified from the original. An earlier version incorrectly stated that 80 percent of the genetic risk factors for autism arise spontaneously. The exact contribution of spontaneous genetic variants to autism is not known.


1: Jacquemont S. et al. Am. J. Hum. Genet. Epub ahead of print (2014) PubMed

2: Levy D. et al. Neuron 70, 886-897 (2011) PubMed


7 responses to “Girls protected from autism, study suggests”

  1. RAJensen says:

    ‘Nearly 80 percent of the genetic risk factors for autism arise spontaneously, or de novo, and these tend to come from the father’.

    Why do genetic and molecular scientists continue to falsly claim that autism is the most heritable of the developmental disorders? Down syndrome is not inherited but nearly 100% of MZ twins are concordant for Down Syndrome while nearly 0% of DZ twins are concordant for Down Syndrome. Calculating a heritabilty estimate based on the difference in concordant rates between MZ and DZ twins would make Down Syndrome the most ‘heritable’ of the developmental disorders. Is it time to junk classical twin study design that has been a huge obstacle to advancing understanding of the so-called autism ‘heritabilty’.

  2. jwright says:

    Dear RAJensen,

    Thank you so much for your comment.

    I would like to apologize, as I mixed up some information. Spontaneous mutations are thought to account for a good proportion of autism risk, but the exact number is unknown. Of these spontaneous variants, 80% come from the father.

    We appreciate your careful reading of


  3. autismparent says:

    What is triggering the expression of these mutations in so many kids? If geneticists were test every individual in the country, I believe they would fidn that we all have imperfect genomes maye some more imperfect than others. So if a mother carries and transmits a mutation (CNV) why does the mother not show autistics symptoms? The difference, I think, is “something(s)” in the environment of that child is triggering the expression of the mutations. Carrying a genetic mutation may indicate increased risk/ predisposition but what is the trigger? I think the conversation about mutations should always recognize environmental triggers. Mutations that pose an increased risk for autism are not the whole story.

  4. LG says:

    Epigenetics have got to play a large part here. I don’t think a single ‘environmental’ risk factor is present to an appreciable extent in a large population or would be powerful enough to ‘trigger’ the expression of these mutations. There’s more to gene expression than the DNA code.

  5. passionlessDrone says:

    Estrogen is neuroprotective in regards to a status of neuroinflammation. Having a state of glial activation during critical developmental timerames has deleterious effects on neurodevelopment.

Leave a Reply

Your email address will not be published. Required fields are marked *