NEW YORK (GenomeWeb News) – A pair of studies appearing online today in Neuron have unearthed new inherited recessive mutations that appear to either contribute to or cause autism spectrum disorder.
For the first of these, researchers from Massachusetts General Hospital, the Broad Institute, Harvard Medical School, and elsewhere did whole-exome sequencing on nearly 1,000 individuals with ASD and about as many unaffected control individuals.
In the process, they found an over-representation of rare, loss-of-function alterations affecting both versions of the same gene. These included homozygous changes that alter each copy of the gene in the same way, as well as so-called heterozygous compound mutations in which two different LoF mutations occur.
In particular, individuals with autism were about twice as likely as unaffected controls to carry inherited recessive LoF mutations that essentially knock out the function of genes, even in genes that normally have low LoF variation rates — a pattern they verified in more than 550 additional ASD cases and around 4,600 controls.
Based on these and other findings, authors of that study proposed that "rare autosomal and X chromosome complete gene knockouts are important inherited risk factors for ASD."
For another Neuron study, meanwhile, an independent team led by investigators at Boston Children's Hospital and elsewhere kicked off its own search for rare, inherited recessive variants in ASD by doing whole-exome sequencing on individuals from autism-affected families in which parents were more closely related to one another than usual, typically due to marriages between first, second, or third cousins.
There, researchers focused on high-penetrance variants, identifying a handful of inherited changes affecting genes better known for their role in Mendelian conditions. While the mutations themselves were less severe than those described in the same genes in the past, they explained, the effects of these mutations appear to be causal for the ASD cases.
Together with the team's subsequent analyses in hundreds more ASD-affected consanguineous and non-consanguineous families, the results hinted that at least a subset of ASD cases share genetic ties to Mendelian disease.
"Basically, you take a multi-system disorder and if you turn down the volume on the severity of the mutation underlying it, then what you get is sort of graded," co-first author Timothy Yu of Boston Children's Hospital told GenomeWeb Daily News. "You don't see all the canonical features of that multi-system disorder."
ASD has long been believed to have both heritable and environmental components. But as authors of the new studies explained, many of the genetic alterations associated with the condition so far involve spontaneous or de novo glitches not found in either parent.
"One mechanism that we would argue has not been adequately explored in sequencing data — and one which we didn't really have a particularly good mechanism of approaching with previous data — is the idea that a recessive or 'two-hit' model might be a way in which some of the heritable component of autism could be explained," Harvard Medical School and the Broad Institute genetics researcher Mark Daly, senior author on one of the studies, told GWDN.
Daly and his colleagues did exome sequencing on 933 unrelated individuals with ASD and 869 ethnically matched, unaffected controls. Most of the exomes were sequenced using Illumina instruments, though study collaborators based at Baylor College of Medicine's sequencing center ran some of the samples on Life Technologies' SOLiD instruments.
For follow-up stages of the study, the team used exome sequence data for parent-child trios described in a study of de novo mutations in ASD that Daly and company published in Nature last April, as well as data on trios from the Simons Simplex Collection. The researchers rounded out their analysis using data from the 1000 Genomes Project and the National Heart, Lung, and Blood Institute's Exome Sequencing Project, or ESP, for controls.
Together, this data turned up LoF mutations affecting both copies of genes that normally have low levels of LoF mutations. In 91 instances, these LoF mutations led to functional 'knockout' of a given gene, study authors explained.
But two-hit knockouts were twice as common in the ASD group, which had 62, compared with the control group, where they saw 29 complete gene knockouts caused by rare LoF mutations.
The researchers also noted that 37 of the knockouts caused by recessive LoF mutations in the ASD cases affected genes shown to have higher-than-usual expression in the brain in past post-mortem studies. In contrast, that was the case for just over a dozen of the complete knockouts picked up in controls.
Still, more research is needed to fully understand whether certain pathways and processes are more often altered by recessive, LoF alterations in ASD, Daly explained.
"Certainly, now that there is this established signal, we're very much interested in drilling down further to see if this points us to specific types of genes or genes that are particularly receptive to mutation," he said.
In the other study, meanwhile, researchers started out focusing on three autism-affected consanguineous families, using a combination of genome-wide linkage analyses and whole-exome sequencing to dig up rare, inherited recessive mutations that appeared to be causal in ASD.
These families were chosen from a much larger group identified by an international network of neurologists, neuropsychologists, psychiatrists, and geneticists known as the Homozygosity Mapping Collaborative for Autism, or HMCA.
The HMCA is more heavily concentrated in the Middle East, where clinicians have been on the lookout for families involving parents who are first, second, or third cousins who have one or more kids with ASD, explained co-first author Maria Chahrour, a genetics researcher affiliated with Boston Children's Hospital and other centers in the area.
When they sorted through genetic data for the families, members of that research team did find homozygous recessive mutations.
A candidate gene identified in one of the families — SYNE1 — had ties to synaptic plasticity. But the inherited recessive mutations segregating with ASD in the other two families involved genes linked to developmental and/or metabolic disorders.
In one of the families, for instance, they found three affected children carrying mutations in a gene called AMT, best known for its role in a neonatal syndrome called nonketotic hyperglycinemia. In the other family, they found changes to the peroxisomal gene PEX7, which causes a metabolic syndrome rhizomelic chondrodysplasia punctata when absent.
Such patterns prompted the team to focus in on a collection of genes behind Mendelian conditions that share features with autism in whole-exome sequence data for up to 163 more large and/or consanguineous families.
Again, that analyses uncovered mutations in genes involved in neuronal processes as well as genes that tend to be more severely mutated in individuals with metabolic or developmental syndromes.
"It turns out … that the neurological manifestations were one of the last things to go away as the severity of the mutation moved from a strong loss-of-function mutation to a mild loss-of-function mutation," Yu explained.
Data for hundreds of parent-child trios or quartets from the Simons Simplex Collection hinted that similar sorts of homozygous or heterozygous compound mutations also appear in a subset of ASD cases found in non-consanguineous families, too.
Together, the study authors said, the work "shows the utility of [whole-exome sequencing] for identifying specific genetic conditions not clinically suspected and the importance of partial loss of gene function in ASDs."