Researchers investigate brain chemistry and structure in neuroimaging project

CHDD research affiliate Dr. Stephen Dager and his colleagues are investigating this question by using sophisticated, non-invasive imaging technology to characterize anatomical features and chemical activity in the brains of young children with autism. The neuroimaging project led by Dager, professor of psychiatry and behavioral sciences and bioengineering, is one of the individual projects in the longitudinal study.

To elucidate the developmental process responsible for structural differences in the brains of children with autism, investigators will image the brains of children participating in the study at two different ages­3 to 4 years and 6 to 7 years. As in the other projects in the longitudinal study, the children whose brains will be imaged include children with autism, children with mental retardation and typically developing youngsters.

Dager's co-investigators include CHDD research affiliates Dr. Kenneth Maravilla, professor of radiology and neurological surgery; Dr. Todd Richards, associate professor of radiology, and Dr. Alan Unis, associate professor of psychiatry and behavioral sciences. Other co-investigators on the project are: Dr. Jill Gardner, research assistant professor of radiology; Dr. Cecil Hayes, associate professor of radiology; Dr. Dennis Shaw, assistant professor of radiology, and Dr. Alan Artru, professor of anesthesiology. Dr. Stefan Posse of the Institute for Medicine in Germany, Dr. Jay Giedd from the National Institute of Mental Health, and Dr. James Brinkley, research associate professor of biostructure at the UW, are consultants on the project.

The neuroimaging data Dager and his colleagues collect will be integrated with the behavioral and cognitive data collected on the same children by other projects in the longitudinal study. This will enable the researchers to characterize relationships between brain anatomy, brain chemistry and the behavioral characteristics of autism, and provide information needed for carefully defining the population for the genetics study.

"We're looking for conclusive physical evidence of brain abnormalities in autism," says Dager. "We hope to link what is going on in the brain, chemically and structurally, with clinical factors and evaluate these findings in relationship to genetic differences."

The researchers will use several morphometric techniques for analyzing magnetic resonance images (MRI) to examine brain structure. These methods, including 3D surface rendering and subregion segmentation, will enable them to look for abnormalities on the surface of the brain, differences in the overall proportions of gray and white matter and subtle anatomical differences in specific regions of interest, such as the cerebellum and hippocampi. To shed light on brain developmental processes in autism, the researchers will assess whether brain structural differences between the group of children with autism and the comparison groups change in the time interval between the first and second evaluations.

"We are trying to understand underlying mechanisms for structural changes over time, whether there's a different growth curve in brain structure and tissue composition in kids with autism versus other kids," says Dager. "And, in conjunction with these very detailed morphometric studies, we want to look at the time course of chemical changes." He expects that the combination of chemical and structural data will yield a powerful tool for delineating clinically different subtypes of autism. To assess chemical changes in specific regions in the brain, Dager and his colleagues will use an innovative metabolic imaging technique called proton echo-planar spectroscopic imaging (PEPSI). "PEPSI gives us the advantage of very fast spectroscopic sampling with good anatomical resolution," explains Dager. Because the PEPSI procedure is done at the same session as the MRI, the method forges a solid link between anatomy and chemistry.

"There has been very little work done trying to understand normal brain chemistry in children and even less in trying to understand abnormal patterns of brain chemistry and changes in brain chemistry with maturation, and how these might relate to autism," Dager notes. "We don't know whether there's going to be a distinct chemical fingerprint for brain abnormalities. Certainly there's reason to believe that if we're finding structural differences we should find corresponding chemical differences. Based on the pattern of those changes, we may begin to understand the underlying mechanisms for the structural changes."

Examples of neuroimaging that will be used to study brain structure and chemistry. High resolution MRIs like the one on the left provide structural information. Four PEPSI images for studying brain chemistry are shown below (along with an example of one of the many spectra that are put together to make each PEPSI image). The PEPSI images portray the same brain location as the MRI. Each image shows the distribution of a specific brain chemical and corresponds to a specific peak on the spectrum.