Subscribe to eTOC

FIRST PURE AUTISM MOUSE MODEL CREATED

ARTICLE IN BRIEF

Mice with a mutated version of the gene for neuroligin-3 demonstrated traditional autistic behaviors — they were less social and showed more repetitive, stereotyped behaviors than wild-type mice.

SALT LAKE CITY—Researchers report that they have developed a genetically accurate animal model of autism that isn't part of a broader neuropsychiatric syndrome.

“The field has been lacking genetic models of pure autism syndromes,” one of the investigators, Craig M. Powell, MD, PhD, said here at the annual meeting of the American Neurological Association.

Dr. Powell, an assistant professor of neurology and psychiatry at the University of Texas Southwestern Medical Center in Dallas, hopes the model will lead to better treatment of the disease.

While some praised the work as valuable, others weren't so sure how well the mouse model will translate into meaningful treatment for people with autism.

Dr. Powell noted that loss-of-function mutations in neuroligins—a family of proteins that act as neuronal cell-surface receptors — and their presynaptic binding partners neurexins have been identified in people with autism or mental retardation (the Fragile X syndrome). In 2003, Thomas Bourgeron and colleagues at the Pasteur Institute in Paris reported in Nature Genetics that they found mutations in two X-lined genes encoding neuroligins in two sibling pairs with autism spectrum disorder. These mutations, they reported, affect cell-adhesion molecules at the synapse and suggest that a defect of synaptogenesis may predispose people to autism.

“We're trying to mimic those mutations in the mouse model, to understand what's wrong with the brain and how that affects the animal's behavior,” Dr. Powell said. “By learning about how the brain is dysfunctional we can then test treatments for the behavioral aspects in the model.”

STUDY METHODS, RESULTS

In the current experiment with mice, the investigators substituted the gene for neuroligin-3 with a mutated version. The mice demonstrated traditional autistic behaviors — they were less social and showed more repetitive, stereotyped behaviors than wild-type mice.

Figure

DR. CRAIG POWELL: “Were trying to mimic those mutations in the mouse model, to understand whats wrong with the brain and how that affects the animals behavior. By learning about how the brain is dysfunctional we can then test treatments for the behavioral aspects in the model.”

The experimental mice also showed enhanced spatial learning, performing better on the Morris Water Maze test, which involves seeing how quickly a mouse inserted into a pool of water finds an escape platform below the surface.

“They have behavioral abnormalities in the social domain versus all other behavioral domains so it is a reasonable behavioral phenotype that is relevant to autism,” Dr. Powell said.

“We've identified changes in cortical circuitry and specifically in enhanced inhibitory synaptic transmission,” Dr. Powell added. He and his research team are expected soon to release findings on the mutant mice missing the gene for neuroligin-1, another protein that helps with synapse connections.

Dr. Powell suggested that future research should focus on specific inhibitory synapse types in mice with the mutated neuroligin 3 gene and try to link that altered synapse function with behavioral problems. Finally, work should be done to examine behaviors and synaptic function in mice deficient in neurexin 1, which teams up with neuroligin 1 to create synapse connections.

EXPERTS COMMENT

Kathleen Clarence-Smith, MD, PhD, who has done research in the autism field and now is a medical consultant with the Stanford Washington Research Group, a public policy think tank in Washington, DC, said the study has potential to bear fruit in the treatment of the disease.

“You need a model first,” she said.

Nancy Minshew, MD, professor of psychiatry and neurology at the University of Pittsburgh, agreed. She said mouse models of autism have value because that they support the idea that the disorder is caused by problems with connections in the brain. And, she said, the models provide clues about the specific mechanisms that link developmental problems with connectivity and specific genes.

Figure

DR. NANCY MINSHEW said mouse models of autism the provide clues about the specific mechanisms that link developmental problems with connectivity and specific genes.

“Those links were all postulated for a long time, but without this very real evidence, did not have meaning to many individuals,” Dr. Minshew wrote in an e-mail to Neurology Today. “Those specialties with the greatest interest and longest commitment to autism view the disorder from a behavioral perspective and treatments are based on behavioral models.”

Dr. Minshew continued: “Genetics and developmental neurobiology seem quite remote to parents who want treatments now. These studies are making this real. There are enough pieces in the puzzle and the work can begin to deliver real benefits, though the full benefit of such work is still a decade or so away.”

But Dr. Manuel Casanova, MD, the Gottfried and Gisela Kolb Endowed Chair in Psychiatry at the University of Louisville, and the associate chair for research there, cautioned that the brains of rodents and humans share “salient differences” in size and connectivity that might limit the ability of autism research in mice to translate to people.

Figure

DR. MANUEL CASANOVA cautioned that the brains of rodents and humans share “salient differences” in size and connectivity that might limit the ability of autism research in mice to translate to people.

He has reported in studies that double bouquet cells — neocortical gamma-aminobutyric acid (GABA) interneurons characterized by the vertical bundling of its axon — play a role in the function of the autistic brain. “Rodents lack double bouquet cells,” he wrote in an e-mail.

“Animal models of autism are easy to find,” Dr. Casanova said. “It is just that they lack explanatory power and ultimately bear no relevance those suffering from the condition.”

He concluded: “I think that most psychiatric conditions fall under the purview of the social sciences because they are uniquely human and can't be modeled in animals. We have already lost a lot of time trying to develop animal models in other conditions such as schizophrenia; now apparently we are moving on to autism.”

REFERENCE

• Jamain S, Quach H, Bourgeron T, et al., for the Paris Autism Research International Sibpair Study. Mutations of the X-linked genes encoding neuroligins NLGN3 and NLGN4 are associated with autism. Nat Genet 2003;34(1):27–29.