Genetics and Autism

Written by Stephen M. Edelson, Ph.D.
Center for the Study of Autism, Salem, Oregon

One of the hottest issues in the field of autism today is the role that genetics may play in the development of autism. There is already evidence to support the view that some forms of autism are genetically based. I will try to summarize the major findings.

Dr. Bernard Rimland, Director of the Autism Research Institute in San Diego, commented in his 1964 book, Infantile Autism, that autism is quite prevalent in both members of a monozygotic (identical) twin pair (born from a single egg that divides) and quite rare in both members of a dizygotic (fraternal) twin pair (born from two separate eggs). Since identical twins come from the same egg, they share 100% of their genes. In contrast, fraternal twins share 50% of their genes, just like any brother and sister. Since 1964, other researchers have replicated the results from Dr. Rimland's survey. Thus, the greater the overlap in genes, the more likely both twins will have autism if one twin does.

A state-wide survey of autism was conducted by U.C.L.A. in the 1980's in the state of Utah. The researchers were able to identify 11 families where the father was diagnosed as having autism. There were a total of 44 offspring from these 11 families, and 25 of the children (over 50%) were also diagnosed as having autism. (Parents and children share approximately 50% of their genes.) The incidence rate in these 11 families is much higher than that for the general population (4.5 out of 10,000 live births). These findings suggest a genetic component of autism. However, one problem with this study was that the person diagnosing the children was also aware that the father had autism.

Many people with Fragile X Syndrome and Rett Syndrome exhibit autistic behaviors (e.g., rocking, social withdrawal, self-stimulatory behaviors). In the former case, the disorder has already been identified as a chromosomal abnormality; and in the latter case, it appears very likely that the disorder is inherited based on the physical similarities of those affected and the fact that all documented cases are female, suggesting a recessive X-linked gene as causing the disorder.

There has also been speculation concerning whether a genetic mutation may cause some types of autism. Genetic mutations occur naturally through the selection and mating of partners (part of Darwin's theory of evolution) or due to the environment (e.g., exposure to radiation or to harmful toxins). It may be the case that a genetic mutation occurred in one person's lifetime, and this mutation was eventually passed down to another generation. Unlike other genetic theories, this theory would be very difficult to support.

Researchers have also proposed that there may be a genetic predisposition to autism. That is, there may be a susceptibility to autism, and an insult to the brain of the fetus, newborn, or infant may cause autism. For example, there is a significant relationship between prenatal maternal exposure to rubella and subsequent autism in the child. However, not every fetus exposed to rubella in utero has autism--some are born blind, some are born mentally retarded, and still others do not suffer any problems at all. One way to conceptualize a genetic predisposition is to envision that a part of the brain is 'fragile;' and when exposed to an insult (e.g., exposure to a virus, lack of oxygen), the person develops autism. An analogy would be being born with a weak arm. If there is no insult to the arm, it will grow strong and there will be no problems in the future. But if the weak arm is damaged or broken early in life, the arm may never function properly.

However, another explanation, which I will call the 'buck shot' theory, can be used to explain how an insult to the brain may not require a genetic predisposition; the insult may still lead to a variety of disorders including autism. If, by chance, the insult to the brain affects a specific area, it may cause autism; however, if it affects another area of the brain, it may not cause autism. For example, if a fetus is exposed to a virus in utero, the virus may attack an area that can cause autism; it may attack an area that can cause blindness; or it may attack an area that does not lead to a disability.

As you can imagine, the possible role of genetics in autism is a complicated issue. However, most professionals are quite confident that someday, we will understand the complex relationship between the role of genetics and the development of autism.