Eric Kandel: A Biology of Mind

Understanding the biology of mental illness would be a paradigm shift in our thinking about mind. It would not only inform us about some of the most devastating diseases of humankind but, because these are diseases of thought and feeling, it would also tell us more about who we are and how we function. I naively thought we were on the verge of such a paradigm change in 1983, when James Gusella and Nancy Wexler were tracking down the gene that causes Huntington's disease. I expected that within 10 years we would have found the major genes that contribute to schizophrenia, depression, and autism. Since then, there has been a lot of enthusiasm about genes and mental illness and some false starts, but surprisingly little progress.

In the past few years, however, certain advances in genetics have given us new reasons for optimism. Now that we can look at the whole human genome, there is a logic to it that we could not appreciate when looking at genes in isolation. As a result, there is reason to believe that the next 10 to 20 years will be more fruitful than the past two decades have been.

One major advance has been the discovery that there is much more variability in the genome than had been anticipated, and that this takes the form of copy number variation (CNV). These are duplications or deletions of segments of a chromosome, often involving several or tens of genes, that enhance or depress the actions of specific genes. A well-known example of a CNV is the extra copy of chromosome 21 resulting in Down syndrome. It has recently been discovered that this type of variation is extremely common in everyone's genome.

A specific type of CNV—called de novo mutations—may be relevant to autism. De novo mutations occur in only one tissue of the body—the sperm or egg—and may crop up relatively late in life (during reproduction), appearing only in the next generation. This fits the pattern of autism, a genetic disease that occasionally emerges in families in which the mother doesn't have autism, the father doesn't have it, and the other sibling doesn't have it. A mother and father could pass this mutation down to one of their children, even though the mutation would not appear in their chromosomes but only in their sperm or eggs. The children would now have the mutation and could pass it on from generation to generation. De novo CNVs may explain the rise in the true incidence of autism in recent years. (Autism cases have also risen in part because of better diagnostic criteria.) This class of mutation, it turns out, is more likely to occur in people who have children late in their 30s and 40s—a segment of the population that has been growing in recent years. Copy number variations and rare de novo mutations may also be a risk factor for schizophrenia.

Scientists are also making progress in finding the biological markers for depression, anxiety, and obsessive-compulsive neurosis. Markers are essential to understanding the anatomical basis of mental disorders, diagnosing them objectively, and following their response to treatment, as well as perhaps preventing psychosis in those at high risk.

The most convincing scientific progress in psychiatry in the past decade has had little to do with genomics. It is the rigorous, scientific verification that certain forms of psychotherapy are effective. This is perhaps not surprising. One of the major insights in the modern biology of learning and memory is that education, experience, and social interactions affect the brain. When you learn something and then remember it for a long time, it's because genes are being turned on and off in certain brain cells, leading to the growth of new synaptic contacts between the nerve cells of the brain. Insofar as psychotherapy works and produces stable, learned changes in behavior, it can cause stable anatomical changes in the brain. We are now beginning to measure such changes with brain imaging. If a person with obsessive-compulsive neurosis or depression undergoes psychotherapy—and if the treatment is successful in changing behavior—the treatment will cause a reversal in the biological markers of these disorders.

Taken together, these advances could open up new approaches to the treatment of depression, bipolar disorders, and schizophrenia, areas that have been at a pharmacological standstill for decades. Along the way we may also learn something about who we are.