The underpinnings of autism are turning out to be even more varied than the disease's diverse manifestations. In four new studies and an analysis published June 8 researchers have added some major landmarks in the complex landscape of the disease, uncovering clues as to why the disease is so much more prevalent in male children and how such varied genetic mutations can lead to similar symptoms.

Large genetic studies have ruled out the idea that the malfunction of a universal gene or set of genes causes autism. And the new papers, which assessed the genomes of about 1,000 families that had only one autistic child, revealed that the genetic mutations that are likely responsible for the disorder are exceedingly rare—sometimes almost unique to an individual patient. Even some of the most common point of mutations were found in only about 1 percent of autistic children.

This finding means that the number of genes lurking behind autism spectrum disorder (ASD) is at least "in the hundreds," says Matthew State of Yale University's Program on Neurogenetics and co-author of one of the new studies. "That's a significant change from the '90s when it was [thought to be] five to 15." And getting a handle on such rare genetic mutations—even in the growing autistic population—is challenging.

Despite the rarity of these genetic code errors, researchers could detect some important patterns in the disparate data. One aberrant gene has already been linked to other social disorders. And by analyzing the role of these genes in neural development, one team of researchers suggests different genetic mutations might often disturb an entire common network.

Down the road, these developments might benefit treatment, too. "It sets the stage to think about it in a new way," says State, whose group's work appears in Neuron.

These large studies are "a good step forward," says Simon Gregory, an associate professor of molecular genetics and microbiology at Duke University, who was not involved in any of the new research. They "enable us to confirm what we'd thought about genetic rearrangements" and are "very important" in having pinpointed new relevant pathways, he notes.

Family patterns
Although autism has been established as a genetically based disease, it does not seem to be passed along in families in the same way that Huntington's disease is. Because those with ASD rarely end up having children of their own, mutations are unlikely to become widespread in populations.

Studies of twins and other families in which more than one child has ASD have shown that autism does have strong genetic roots, but the new studies sought to get past the commonalities and search instead families in which only one child has the disease.

"You see clearly that if you compare the autistic kids with their [unaffected] siblings, they have more of these mutations," says Dennis Vitkup of the Department of Biomedical Informatics at Columbia University and co-author of one of the new studies published in Neuron.

In assessing such a large and diverse data set several of the studies all alighted on a genetic explanation for one of the most striking patterns in ADS: why at least four times as many boys than girls are diagnosed with the disease.

Girls, it seems, might better resist the development of autistic signs: Bigger genetic disruptions are necessary to cause ASD to manifest in girls than in boys, according to the new analyses. Girls might be better protected against autism-causing genetic anomalies, Vitkup suggests, because they tend to have stronger social inclinations than boys.

Although the ability of girls to withstand genetic mayhem might seem to predispose them to become silent carriers of autism, the new analysis shows that mothers were no more likely than fathers to pass on harmful mutations.

Social genes
To decipher the code of autism, researchers are also looking outside of the ASD patient community to other developmental and social disorders.

One of the few rare mutations that cropped up in some autistic children in the studies were extra copies of 7q11.23 (shorthand for denoting the positions, or loci, of the genes on the chromosome—in this case on the long arm, or "q," of chromosome 7). As several of the research teams pointed out, deletion of this region has been implicated in Williams-Beuren syndrome (WBS), a disease that tends to make people especially gregarious, empathetic and social.

"There's clearly something in that small region—of 25 or so genes—that's having a significant impact of social interactions," State says. "Within that relatively very small region in the genome there are going to be keys to studying neurology and social development."

Mutations at other regions of the genome did crop up more than once in the study group. And a copy error at 7q11.23 or other loci did not necessarily translate into similar levels of IQ or developmental disability in different patients. Hence, factors other than errors at these loci must also be playing a role in the manifestation of ASD.

Rather than wait for additional genome scans to turn up more potential mutations, however, many research teams are already creating models of how these mutations might impact neurological development.

Although such model building might seem premature given the ever-changing genetic terrain of the disease, "having a way to begin to interrelate them might actually help to study them," says Huda Zoghbi, of the Department of Molecular and Human Genetics of Baylor College of Medicine, who co-authored an analysis of the three Neuron studies published online today. So rather than get mired in finding each possible gene, she says, it makes sense to "go back and forth between the genetics and the functional studies."

Finding the function—and dysfunction
Vitkup and his team conducted just such a functional, model-based approach. Their paper, published online in the same issue of Neuron, looked closely at the location and likely effects of the mutations among families that have only one ASD child. By figuring out which genes communicate with each other, he says, you can "see if mutations try to disrupt genes that are next to each other," and thus what common pathway different mutations might be messing up. He likens it to a hunt for a criminal that might be committing robberies in different states but with the same modus operandi, perhaps choosing similar targets each time.

With a little computer-assisted detective work he and his team found one cluster of pathways that many of the errant genes seemed to be interrupting. And it turned out to be a crucial cluster, involved in synaptic development and the movement of neurons in the young, developing brain. As neurons branch out to form connections, if some pathways are disrupted, the connections can become abnormal. In a sample of about a dozen cases, Vitkup says, most of the patients had disruptions that would encourage an overabundance of particular neuronal connections. Such a pattern provides evidence for the excess of connections in autistic children producing the opposite behavior pattern from WBS, whose patients have fewer than normal connections. But, he says, the jump from genetic mutations to social skills is difficult.

Nevertheless, that mutations implicated in ASD would be linked to this sort of neuronal network "is logical by the phenotype," Vitkup says. And for future studies and diagnoses, he says, it "can help because we can now look to see if there is a new mutation somewhere in the genome and we can see how close—or how related—the new mutation is to our cluster."

He and his team currently have several dozen genes mapped into their network, but he expects the list to grow to as many as 500 in the next few years as more individuals with ASD are included in these studies and as sequencing technology improves. And there might well turn out to be other key clusters of pathways that are discovered, which will have an entirely different list of implicated genetic mutations, Vitkup says.

Zoghbi and her team, whose work published online June 8 in Science Translational Medicine, have gone through much of the same data to find patterns in the types of proteins that these rare mutations might be affecting. A new genetic mutation can change the way proteins are made—they might be made incorrectly, too often or not at all. "This can have a domino effect on many other proteins that could affect how a neuron talks to another neuron," Zoghbi explains. She likens it to a self-contained neighborhood in which each person has a particular skill set. If everyone is present and working well together, the garbage will get collected and the streetlights will stay on. But if one or two people are missing or unable to do their work properly, major systems will start to falter, "because the other ones don't have those skills," she says. Likewise, "a group of proteins is needed for a cell to function well."

With just a couple dozen proteins flagged a few years ago, Zoghbi and her team now have hundreds that they have added to the growing list of autism instigators.

"The more we understand the function of the proteins involved in autism—and by what pathways they might impart that change—we might begin to ask, 'Where can we intervene, and would one intervention help just one patient or a group?'"

Screening and treatment
By better understanding the numerous routes autism can take to perturbing common pathways, new avenues of treatment might open up sooner. Currently, treatment is based on behavior or serendipity, State says, adding, "we're very far behind other areas of medicine in that respect." But, he says, if a genetic screen can find even a rare mutation in a child before symptoms appear—or even in utero—behavioral therapy could start earlier, improving that child's level of functioning.

And for pharmaceutical development, if treatments can be pinpointed to improving a common pathway, rather than fixing a particular genetic error, they might be able to treat a wider range of ASD patients instead of each individual type of mutation. But interventions like these are "easier said than done," Zoghbi notes. "There are lots of proteins involved and lots of genes involved."

Genetics are, of course, just part of the increasingly complex autism puzzle. "Two people can have exactly the same mutation" and not have the same degree of developmental disorder, State says. "The question of why is the multimillion-dollar question."

To help sort out this increasingly urgent answer, Gregory advocates for a broad-spectrum approach. "It's not going to be one thing, it's going to be a collection," he says. "Between genetic, genomic and epigenetic, we'll identify what causes the spectrum." (Epigenetics refers to the environmental modification of genetic activity; such changes can be heritable.) And within these, the environment is often another complicating factor, as a person's genetic makeup can render them more or less sensitive to environmental influences—whether that is from social bonding or purported chemical influences.

But one thing is well established in autism research: as scientists look deeper into the disease the complexities multiply almost exponentially. Gregory suggests that one of the next steps will be to assess the mechanisms behind epigenetic influences in autism. But "that becomes a harder thing to answer," he says, speaking from experience in that field. DNA methylation and its effect on genes varies in different types of tissue, adding another layer of challenge to parsing the interdependent effects.

The other research teams are also hard at work on the next batch of studies. State's group is expanding their study to include some 1,600 more families as well as homing in on gene regions that they have already found.

The rush of studies in the past couple years has been thanks in large part to technological advances as well as a push to study the disease more closely. "The down payment in the early part of this century is really paying off," State says. But Gregory is eagerly anticipating "the next big leap forward" in higher-resolution sequencing, which will allow is group and others to "identify these very small changes" that researchers are now only just getting a taste of.