Thomas Insel, M.D., says that a "midterm goal" of the
revolution now under way in psychiatry is the development of"
biodiagnostics" to identify biomarkers for specific mental
illnesses. David Hathcox
Breakneck advances in genomics promise to usher in an era of"
individualized medicine" based on a thorough understanding of the
molecular pathophysiology of mental illness.
So said Thomas Insel, M.D., director of the National Institute of Mental
Health, in the Judd Marmor Award Lecture at APA's 2005 annual meeting in May
Insel sketched a portrait of a not-too-distant future when clinicians will
treat precise targets along a pathophysiological chain from genes to cells to
distributive systems within the brain based on a patient's unique genetic
This is an ambitious, even visionary, portrait, but one that Insel said is
already taking shape in the treatment of cancer, diabetes, and cardiovascular
disease. While this vision depends on some technological virtuosity yet to be
attained in psychiatric research, he noted that these advances are already
being pursued with "breakneck speed."
For instance, merely knowing the location of genes, as was achieved with
the completion of the Human Genome Project, is barely a beginning. That
accomplishment has been likened to writing the "White Pages," a"
text" made up of 3 billion base-pairs of DNA, with every gene
having an address and a phone number to locate it within the text.
But what is really necessary, Insel explained, is the "Yellow
Pages"—a catalog of where and how genes are expressed and how they
function. So a critical research goal is to go "gene by gene along the
White Pages, ask if the gene is expressed in the brain, and if so
That goal is being advanced by the Gene Expression Nervous System Atlas
(GEN-SAT) Initiative at Rockefeller University, among other places.
And that is not all. Insel noted that scientists possess a"
consensus" genome sequence derived from the handful of people who
contributed their DNA to the public and private arms of the Genome
But, like the proverbial snowflake, no two people who are not identical
twins will ever have the exact same genome. What is really necessary to close
the link between the genome and human health or disease is a map of variations
across the 3 billion base-pairs of DNA.
"That is the challenge," Insel said, "making the leap
between genomic variation and behavioral or functional variation at the level
Fortunately, variations in genomic sequence (known as single nucleotide
polymorphisms) occur in inherited units known as "haploid
genotypes," or haplotypes— meaning that scientists do not have to
map all 3 billion base-pairs of DNA for variations; they only have to map the
Spearheading this project is the International HapMap Project, a
multicountry effort to identify and catalog haplotypes. The next version of
the map is scheduled to be completed this summer, Insel said.
"This tells us for the first time that we can begin to study
individual variation at a level and a speed and at a lower expense than we
ever thought possible," he said. "The question of how you relate
individual variation in sequence to function is now a tractable
The upshot of these developments is likely to be transformative.
"Where this may take us is to a very different vision of what
psychiatry could look like in a postgenomic era," Insel said. "We
are talking about moving us from where we currently diagnose by symptoms and
treat empirically to an era where we really do understand something about the
molecular pathophysiology of [psychiatric] disorders."
Insel predicted that the current and long-standing strategy by which
treatments for mental illness are derived will be looked back on as
anachronistic—and somewhat illogical.
Today, he said, the pathophysiology of mental illness is surmised from the
mechanism of action of pharmaceutical compounds that are themselves chanced
upon serendipitously, as a result of tweaking other formulations already
It is a process that stands scientific logic on its head. "It's like
trying to find out if the mechanism of aspirin is related to the
pathophysiology of headache," Insel said. "Thinking that way is
not going to lead you to new discoveries, but to a lot of knock-offs of
aspirin. Yet that is very much what has happened in psychiatric research. We
haven't had any new classes of compounds in 30 years."
The modern molecular genomic model reverses this paradigm. Already, in
areas of research such as cancer, diabetes, and cardiovascular disease,
scientists are elucidating specific variations in genetic sequence that result
in alterations in cellular and systems functioning; then animal studies are
designed to test targets along the chain from genes, to cells, to systems, to
"That's the way you get to new classes of drugs—not by taking
drugs we already have and modifying them to find a market," he said.
In psychiatry, the task is complicated by the fact that all of the
psychiatric conditions are considered "complex" disorders
involving multiple genes and multiple brain systems.
The way in which the genomic text itself is "read out" in
cells, systems, and functioning is proving more baroque than previously
imagined. One curiosity emerging from the completion of the Human Genome
Project, for instance, is the relatively small number of genes; of 3 billion
base-pairs of DNA, there are just 23,000 genes, or intelligible"
sentences" in the text.
A question this raises is the nature and purpose of the great bulk of
nongenetic material in the genome, and Insel said much recent research is
focused on these noncoding areas: What are they there for? And what are they
Research within the last year reveals that it is in the process of
transcription (by which an RNA copy is built from a DNA sequence) and
translation (the process by which the RNA copy is translated into the amino
acid sequence of a protein) that these noncoding regions play a critical
For instance, "promoter" regions of DNA appear to influence
where and how much of a protein is read out, so that a variation in a promoter
region can have an enormous influence on cell and system function, he
For this reason, Insel said it is the complex processes of transcription
and translation—not merely the gene itself—that holds the key to
understanding how the genetic blueprint expresses itself in cells, systems,
Some recent and ongoing research is managing to put all the pieces of this
puzzle together, a model for a true molecular genomic understanding of mental
illness (see box).
Insel suggested that a midterm goal of this revolution-in-progress would be
the development of "biodiagnostics" using, for instance,
neuroimaging tools as biomarkers for specific mental illness.
"What we are talking about is developing treatments that go after the
core pathology," Insel said.
But the real breadth of his vision for the revolution in molecular
neurobiology was suggested in his comments about the"
endgame"—when psychiatrists could use knowledge of
individual genomic variation to prevent and cure major mental disorders,
rather than merely manage and treat them.
"The endgame is individualized care," Insel said. "It is
not hard to imagine a time when you would know which person was at very high
risk for schizophrenia, and you would have the neuroimaging data you would
need to watch a patient very carefully. At age 16 when the patient begins to
develop a sleep disturbance and ideas that are even stranger than those of his
peers, you would know that this was a patient with whom you would want to
He asked, "Why aren't we thinking about preventing the first break
and putting together the kind of science that we need to do that?"
Information about the International HapMap Project is posted online