Annual Report [2018]

SFARI Research Roundup

Autism Research (SFARI)

Since the launch of the Simons Foundation Autism Research Initiative (SFARI) in 2003, the initiative has supported nearly 500 investigators. In 2018, SFARI Investigators studied a wide range of topics — explicating the role of missense mutations in autism, identifying a biomarker for low sociability in monkeys, and understanding how an autism-linked mutation affects brain wiring, for example. Here is some of the work of SFARI Investigators over the past year. 

Sociability Boost. For decades, researchers have suspected a link between brain serotonin levels and autism. Yet trials of serotonin-increasing antidepressants as treatments for autism have proved disappointing. An August 8, 2018, study in Nature suggests a reason why: These drugs may not target the right brain pathway with enough specificity.

A team led by SFARI Investigator Robert Malenka of Stanford University used light to activate a particular brain pathway in mice connecting the serotonin-producing dorsal raphe nucleus to particular serotonin receptors in the nucleus accumbens, a brain region involved in rewards. Activating the pathway, the researchers found, temporarily made mice far more sociable. 

And in a mouse model of autism in which neurons in this pathway lacked the autism-linked chromosomal region 16p11.2, the team again found that activating the pathway made the mice more sociable, ‘rescuing’ them from the effect of the 16p11.2 deletion. Malenka’s team is now studying whether drugs that activate serotonin receptors directly might be more successful than previously studied antidepressants as treatments for mouse models of autism.

A genetic mutation altering just a single amino acid in a protein can affect how that protein interacts with other proteins. This illustration shows the protein interaction interface between two proteins, TRIO(blue)and RAC1 (yellow). The orange spheres represent mutations linked to intellectual disability and microcephaly, while the red sphere represents a mutation associated with autism spectrum disorder. S. Chen et al./Nature Genetics 2018

A Neural Pacemaker. Even when a case of autism springs from a clearly identified genetic mutation, there’s a huge gap between understanding which gene is malfunctioning and repairing the damage it has caused. A new study suggests that it may not always be necessary to make this leap to treat autism. Instead, it might be possible to develop a treatment analogous to a cardiac pacemaker, which helps heart cells coordinate better instead of repairing them.

A team led by SFARI Investigator André Fenton of New York University recorded the electrical activity of neurons in the brains of mice with the autism-related fragile X syndrome. The researchers then observed the mice’s “place cells” — cells in the hippocampus that keep track of where the mouse is — as the mice performed a task that involved adjusting to changing locations, something mice and people with fragile X syndrome struggle with.

The researchers reported in the February 7, 2018, Neuron that the mice’s place cells individually processed spatial information normally but did not coordinate well with each other, often failing to form the temporary coalitions needed to perform cognitive tasks. This result suggests, Fenton said, that neuromodulation, an emerging array of techniques that apply electrical pulses to coordinate brain activity, might be useful for treating this lack of adaptability.

Sociability Marker. Autism researchers have long looked for a stable biomarker of autism — something measurable in a person’s bodily fluids or tissues that correlates with autism symptoms. Studies of blood have mostly come up empty, but a new study suggests a potential biomarker for social deficits in cerebrospinal fluid: low levels of the molecule vasopressin.

A new study suggests a potential biomarker for social deficits in cerebrospinal fluid: low levels of the molecule vasopressin.

SFARI Investigator Karen Parker of Stanford University and her colleagues (including SFARI Investigators Antonio Hardan of Stanford and Elliott Sherr of the University of California, San Francisco) reported in the May 2, 2018, Science Translational Medicine that in naturally occurring rhesus monkey populations, the least sociable monkeys had markedly lower vasopressin levels in their cerebrospinal fluid than the most sociable monkeys did.

Compared with rodent models, monkey models of autism offer an especially promising way to study the disorder, since primates are so much closer to humans. And indeed, in a small human study, Parker’s team found that children with autism also had significantly lower vasopressin levels in their cerebrospinal fluid than controls did. Parker and her colleagues have started clinical trials of inhaled vasopressin as a treatment for low sociability in autism, with encouraging preliminary results.

From Mutation to Miswiring. Recent studies have made enormous strides toward identifying the mutations that underlie autism. But these genetic variants confer autism risk through a wide variety of mechanisms, and little is known about how most of these mutations affect brain connectivity and function.

A new study illuminates this question for one of the most common genetic causes of autism: deletions in 16p11.2. SFARI Investigator Alessandro Gozzi of the Italian Institute of Technology in Genoa and his colleagues examined brain imaging data from children with 16p11.2 deletions in the Simons Searchlight cohort. They report in the July 1, 2018, Brain that the children have impaired connections between the prefrontal cortex, a brain region involved in social behavior and cognition, and other brain regions. This weakened connectivity correlates with low social and cognitive skills.

The team also found that mice with a 16p11.2 deletion have flawed connections between the prefrontal cortex and the retrosplenial cortex, which is involved in cognitive functioning. These mice showed a wide suite of brain impairments in the prefrontal cortex, such as miswiring of the neuronal projections connecting the region to the thalamus and a shortage of dendritic spines. The findings suggest that this mouse model may offer a faithful window into the neurobiology of people with 16p11.2 deletions.

Prioritizing Missense. Missense mutations, in which only one amino acid in a protein gets altered, are thought to underlie many cases of autism. Thousands of such mutations have been found among children with autism, but it is hard to know which of these mutations disrupt the function of the gene’s corresponding protein. A new framework for prioritizing missense mutations, published on June 11, 2018, in Nature Genetics, aims to change that.

The study — led by SFARI Investigator Haiyuan Yu of Cornell University, along with SFARI Investigators Bernie Devlin of the University of Pittsburgh and Kathryn Roeder of Carnegie Mellon University — looked at thousands of missense mutations in children with autism and their unaffected siblings in the Simons Simplex Collection. The researchers found that not only were the children with autism significantly more likely to have a missense mutation than their siblings, but these mutations (unlike their siblings’) were 27 percent more likely than chance would predict to affect a region that interacts with other proteins.

By combining experimental methods with machine-learning tech-niques, the team found that missense mutations resulted in about 2.5 times as many disrupted protein interactions in children with autism as in their siblings. And these disruptive mutations were more likely to impact ‘hubs’ — proteins that interact with multiple other proteins — than in the siblings. This protein interaction framework, the researchers argue, offers a novel way to identify which missense mutations are most likely to confer autism risk.