Current Issue
Heart Heal Thyself
A Chip Off the Old Genomic Block
Fighting Infant Mortality
More than a Medicine Bottle Dean's Message
In Brief
Gifts and Support
Class Notes
Past Issues
Contact Us
Other Publications
Give a Gift
School of Medicine




E-mail to a Friend     Printer friendly
By Sylvia Wrobel  
BASED ON TWIN AND FAMILY STUDIES,autism is one of the most highly heritable of all neuropsychiatric disorders. Its complex genetics likely involve the interaction of multiple susceptibility genes, acting in concert with one or more unknown environmental factors. Although few of these susceptibility genes have been identified, Emory human geneticist Michael Zwick believes he knows where more are hiding.
     Zwick focuses on a narrow region of the X chromosome for two reasons. First, because boys have only one X chromosome, they are a least four times more likely than girls (who have two X chromosomes) to be affected by autism. Second, the X chromosome neighborhood already is known to be home to a large number of genes that cause problems with brain development when severely mutated. That neighborhood includes the fMR1 gene responsible for fragile X syndrome, the most common cause of inherited mental retardation. One of five children with fragile X also meets the criteria for autism, making fMR1 the most common known cause of autism, even though it accounts for only 2% to 5% of cases.
     However, Zwick believes the relatively small number of cases may be only the beginning of the fMR1/autism story. Fragile X is caused by a triple repeat in the fMR1 gene, an error that inactivates the gene. Zwick's boss, Stephen Warren, chair of Emory's Department of Human Genetics and discoverer of the fragile X gene, previously found another mutation, unrelated to the classic triplet repeat, in the fMR1 coding sequence. Numerous such variations may occur in fMR1 and in the nearby fMR2 gene, also known to cause mental retardation. But until Zwick's efforts, no one had looked at the nucleotide makeup of the fMR1 gene where disease-causing variations could be spotted. There was a good reason. Only recently, thanks to work by Zwick and a handful of other pioneers, has the technology existed to search efficiently for such subtle variations.
Dime-sized technology
The project to sequence the human genome was necessarily enormous, deliberately anonymous and generic, and prohibitively expensive, restricted to large "industrial-strength" facilities. Zwick envisions a more practical alternative to genome sequencing that is focused, individualized, fast, and cheap. That is why he is developing technology and software that enable smaller laboratories to rapidly and inexpensively generate large quantities of genetic data for individuals.
     One surprise in sequencing the human genome was how few gene differences are found among individuals. On average, any two human genomes have only eight differences for every 10,000 base pairs (the order of A-T-G-C nucleotides that determines which protein a specific gene will produce).
     Resequencing compares the genome of individuals or groups (such as patients with a specific disease) to the entire human genome to identify where those differences occur and to find out if a larger than expected number of people with the same condition have the same differences.
     In his search for autism susceptibility genes, Zwick takes this process one step further by looking at how the genes of boys affected by autism and their healthy fathers differ from both the human genome and from each other. His research, conducted in a pantry-sized area and using equipment that would easily fit in the trunk of a car, is a good example of the emerging technology's power and efficiency.
     While Zwick often speaks to families touched by autism, he does not know the autistic sons and their fathers whose DNA fills his dime-sized resequencing chips. The patient samples and information come from the Autism Genetic Resource Exchange, a gene bank with pedigrees, genomic scans, and/or DNA samples of almost 830 families with more than one member diagnosed with some form of autism.
     From this wealth of data, Zwick first selected 314 families with two affected sons, both of whom share the same portion of the X chromosome from their mother. He then chose one of each pair of brothers at random and sequenced the region of his chromosome containing the fMR1 and fMR2 genes, as well as all the unique noncoding sequences outside these genes. He next sequenced the same region of the X chromosome for each father of the boys with autism.
The project to sequence the human genome was necessarily enormous, deliberately anonymous, and prohibitively expensive, restricted to large ‘‘industrial-strength" facilities. Zwick envisions a more practical alternative to genome sequencing that is focused, individualized, fast, and cheap.
Illuminating father-son variants
Before this genetic information is resequenced, first against the human genome, next against each father-son pair, each male's genetic material is hybridized—a process of mixing and separating the DNA in liquid form to yield paired DNA fragments—on a computer chip three-quarters of an inch square, no thicker than a microscope slide. A laser scans the chip to visualize the male's DNA and determine its sequence.
     The chip with genetic material from the son with autism and the chip with genetic material from his father then are entered onto a computer loaded with software previously developed by Zwick and his colleagues at Johns Hopkins. The screen fills with an array of colored dots, and the bright lights mark the genetic differences between the son and father.
     Variants from the human genome found in both individuals probably do not cause disease—or at least are not related to autism. Variants from the human genome found only in the son, and not in his father, are candidates for autism susceptibility gene mutations. Statistical analysis of all 314 father/son sets will determine how common, how rare, and how meaningful such differences are in relation to autism. The cost of using the new technology? Less than .001 cent per base pair.
     Identifying variations contributing to autism would help scientists develop new early diagnostic tests, leading to earlier intervention. Understanding which genes—or more precisely which specific mutation or mutations of specific genes—are involved in susceptibility to autism would guide development of treatment, perhaps even lead to the first drug therapy. Knowing which genes are involved would also elucidate environmental influences that may contribute to autism.
     Even if he finds the variations he is searching for, Zwick knows they will be only part of the complex autism puzzle. But his new technology stands ready to be used to look for more answers throughout the genome, and its implications go beyond any one disorder. Next-generation genomics technologies like Zwick's will help deliver a genome-sequencing center on every laboratory bench.
Copyright © Emory University, 2007. All Rights Reserved