Autism Biomedical Information Network

Scientific American, February 2000, The Early Origins of Autism

Summary and Commentary

Although there is little doubt that autism has a neurobiological basis, the timing of the onset of the neurodevelopmental aberration has not been clear. Most researchers believe that some type of insult occurs in utero during a critical phase of brain development. On the other hand, some parents of children with autistic regression have had difficulty accepting the possibility of an in utero event and believe that some type of postnatal insult, perhaps an environmental toxin, affected their childrens' brain development.

The article, The Early Origins of Autism (Scientific American, February 2000, p. 56-63), provides additional evidence for an antenatal event occurring quite early in pregnancy. Dr. Patricia Rodier, an embryologist at the University of Rochester, was stimulated by a report in 1994 of a relatively high incidence of autism (about 5 percent, which is about 30 times higher than found in the general population) in adults whose mothers had taken thalidomide during the early weeks of pregnancy in the 1960s. Many had the range of limb anomalies expected with in utero exposure to thalidomide. Since the teratogenic effects of thalidomide occur quite early, between 20 and 36 days after conception, this observation suggested that the brains of affected persons with autism might have suffered similar serious injury during embryonic development. Furthermore, most of the thalidomide-affected adults with autism also had subtle deformities of the external ear. The ear anomalies suggested that the "window" of injury occurred between 20 and 24 days (the period of differentiation of the external ear structure). This period also corresponds to the development of the brain stem, especially nerve cell clusters that control the muscles for eye movement and facial expression.

Rodier and colleagues then studied the preserved brain tissue of a young woman who had died many years before and found marked attenuation in the size of the facial nerve nucleus (important for the muscles of facial expression) and the superior olive (involved in auditory processing). Moreover, Rodier recognized that the foreshortening of the length of this segment of the brain stem resembled that found in mice with a "knockout" of the HOXA1 gene. This gene, found on chromosome 7, plays a critical role in the earliest stages of embryonic neurodevelopment but is "switched off" later. It has two protein-coding regions and is responsible for the production of a transcription factor modulating the activity of other genes.

Rodier et al then studied this gene in humans and found that a specific polymorphism (a variation in the sequence of nucleotide bases in DNA) occurs twice as frequently in autism compared to individuals without autism (40 percent versus 20 percent). This means that 60 percent of persons with autism lack this variation and must have other, yet-to-be-defined genetic factors. This is not surprising since it is believed that several susceptibility genes must co-occur for autism to occur as defined by DSM-IV criteria. The interaction of fewer susceptibility genes probably accounts for the "broader phenotypes" in the autistic spectrum (such as PDDNOS or Asperger's disorder), or even apparently-typical development in close relatives but with subtle indications of social maladroitness.

Since there are a number of other autism "hot spots" in the genome, studies of other genes involved in brain development are currently under way. Rodier and colleagues also looked at the HOXB1 gene on chromosome 17 but did not find a strong association with autism. Since up to one-half of the human genome of about 100,000 genes may be dedicated to brain development, there are myriad possibilities for mutant "neurogenes."

This evidence of the antenatal origin of autism in the earliest stages of brain stem development, between 3 and 4 weeks after conception (when many women are not yet certain of being pregnant), raises the possibility of exposure to an environmental toxin during this sensitive period. Of course persons with autism don't simply have problems with the muscles of facial expression or with auditory processing. However this early period of embryonic development may be crucial for tying together all the brain mechanisms for social cognition, including the "binding process" that confers the feeling of pleasure of being in the company of others.

It is obvious that the teratogenicity of thalidomide has broad implications for biology and there has been a surge of interest in this compound. One can postulate that a mutant HOXA1 gene and the teratogenic effects of thalidomide on the brain act through a "final common pathway." There is much current research on thalidomide including a recent paper suggesting an anti-angiogenesis effect mediated by reactive hydroxyl radicals. Impaired formation of a protocapillary network in the embryonic brain would have profound effects on development.

Dr. Rodier emphasizes that a better understanding of the neurobiological basis of the early embryonic origin of autism will ultimately provide a rational basis for treatment and genetic counseling.

Full text of Rodier article (with confusing layout of 3-column format interrupted by sidebars)
[NEW LINK 10/24/2003]

Return to list of commentaries and other autism links

Commentary by Ronald J. Kallen, M.D., ©2000, This page last updated on 10/24/03