Autism May Be Caused By An Immune System Response To A Virus

The researchers found that autistic children who had been exposed to certain viruses in the past showed unusually high levels of antibodies to brain proteins, suggesting an autoimmune response. Their findings appear in the October issue of the peer-reviewed journal, Clinical Immunology and Immunopathology.

Autism is a developmental disorder that affects brain function, interfering with reasoning ability, imagination, communication, and social interaction. Children with autism start talking later than other children, and when they do speak, their communication skills are extremely limited. They often avoid looking at other people and don't learn to read others' faces for signs of emotion or other cues. These children typically are unable to play creatively, and some engage in repetitive, sometimes self-destructive, behavior, such as rocking, hand flapping or head-banging.

No single cause of autism has been found, and researchers believe that genes and environmental factors (such as viruses or chemicals) both may contribute. The kinds of brain abnormalities found in people with autism suggest that the disorder arises when something disrupts normal brain development.

One possibility is that early exposure to a virus prods the body into mounting an immune response that somehow goes awry. In addition to producing antibodies against the virus, the body makes antibodies against itself, resulting in damage to tissues and organs. This "autoimmune" response is what happens in autoimmune diseases such as lupus, and some researchers think a similar response may account for the brain abnormalities found in people with autism.

It was this possibility that U-M researchers Vijendra Singh and Victor Yang and undergraduate student assistant Sheren Lin investigated. In their study of 48 autistic children and 34 normal children and adults, the researchers measured levels of antibodies to two viruses---measles virus and human herpesvirus-6---in the subjects' blood. These antibodies were chosen because they are often used in research on known autoimmune diseases, says Singh, the principal investigator of the project and an assistant research scientist in the College of Pharmacy.

The researchers also measured levels of two brain autoantibodies (antibodies to brain tissue). One, anti-MBP, is an antibody to myelin basic protein, a protein found in the protective sheaths around nerve fibers in the brain. The other, anti-NAFP, is an antibody to neuron-axon filament protein, a protein that makes up the nerve fibers themselves.

Virus antibody levels were essentially the same in autistic and non-autistic subjects, as the researchers expected. But the majority of autistic children who had virus antibodies also had brain autoantibodies. The higher the level of virus antibodies, the more likely an autistic child was to have brain autoantibodies. None of the non-autistic subjects had brain autoantibodies.

The strongest link found in the autistic children was between measles virus antibodies and anti-MBP, suggesting that exposure to the measles virus may trigger an autoimmune response that interferes with the development of myelin, says Singh. If myelin in the brain doesn't develop properly, nerve fibers won't work as they should. This could be one way that the brain abnormalities associated with autism arise.

The question of how exposure to measles virus occurs raises a controversial issue. Parents of children with autism often report that the children started showing signs of the disorder shortly after being immunized with measles-mumps-rubella (MMR) or diphtheria-pertussis-tetanus (DPT) vaccine, but no scientific studies have shown a link between vaccines and autism. In the U-M study, almost all the subjects had had MMR immunizations, and none had ever had a case of measles. It is possible, however, that some might have been infected with measles virus but never developed symptoms of measles, says Singh.