Hopkins researchers says maternal antibodies may be linked to autism

From the Johns Hopkins News-Letter, the university's student newspaper:

When you think of "tolerating" a baby, images of enduring sleepless nights, changing diapers and wiping mashed sweet potatoes off the ceiling usually come to mind. However, when a woman becomes pregnant, the idea of fetus toleration is taken to a whole new level. Our immune systems are designed so that our bodies produce antibodies that attack anything that is foreign, such as bacteria, viruses or tissues.

But for most expecting mothers, the fetus, which is indeed foreign because of the presence of the father's DNA, is not rejected. The mother's immune system compensates so that the fetus is not attacked by antibodies and can be carried to term.

Unfortunately, sometimes this fetus-protection immune response goes awry. Hopkins researchers previously found antibodies that target the brain tissue of fetuses in mothers of children with autism.

Autism is a neurological disorder characterized by impaired social interactions, difficulties communicating with others and unusual or repetitive behaviors, the direct causes of which are still unknown.

This finding has caused this same team, led by Harvey Singer, director of Pediatric Neurology, to test a promising but not entirely well-evidenced theory: Certain antibodies, which are passed from mother to child across the placental wall, attack the fetus's brain itself, and may lend a hand in the development of autism.

In a paper appearing in this month's issue of the Journal of Neuroimmunology, Singer and his colleagues determined that these autism-positive antibodies are sufficient to produce at least some of autism's characteristic behavioral patterns in induced-autism mouse models.

When injected with antibodies from mothers with autistic children, expectant mouse mothers had pups that exhibited increased levels of anxious behavior, hyperactivity, increased ease by which the animal could be startled by loud noises and decreased sociability.

On the other hand, for expectant moms that were not injected with antibodies, or that were injected with antibodies from mothers without autistic children, the mouse pups behaved much more normally compared to the experimental group.

Not only did the team discover that the behavioral changes were more pronounced in the mice injected with the antibodies of the mothers of autistic children, but the team also found that the course of the disorder's development followed a similar pattern to that seen in the human disorder: Symptoms intensified as the mice aged. This is similar to the clinical progression of autism, in which a child will develop normally until a certain point, when regression usually begins to occur.

Unfortunately, it is overly tempting to assume that these antibodies are the direct cause of autism-spectrum disorders, just because they seemingly contribute to the development of certain behavioral patterns.

However, this is probably not the case. In a more likely scenario, autism is the result of several confluent factors, including but not limited to abnormal immune responses, environmental factors, genetic influences etc.

This finding does, however, further demonstrate that certain maternal antibodies do cross the placenta into the baby's environment, and that these antibodies seem to play a role in the development of the disease, possibly through induced inflammation of the brain and subsequent cell death due to a faulty immune response.

Further research is needed to determine which proteins or brain regions are affected by these antibodies, as well as to see whether these antibodies can be blocked. This may eventually lead to the ability to inhibit their transfer from mother to offspring, decreasing the risk of the child developing an autism-spectrum disorder.