September 21, 2006

Schizophrenia Risk Genes - Interview with Dr. Daniel Weinberger, NIMH

Interview of Daniel Weinberger, MD, Senior Investigator, National Institute of Mental Health
by Demian Rose, MD, PhD
12th Annual Staglin Music Festival for Mental Health
September 16th, 2006

Background: Dr. Weinberger is the Chief of the Clinical Brain Disorders Branch of the NIMH in 1987. His research lab focuses on basic neurobiological and genetic mechanisms of neuropsychiatric disorders, especially schizophrenia.

You’ve stated that, in your opinion, the genetics of schizophrenia have recently advanced quite rapidly. Can you explain to the audience at why you feel this way?

The search for schizophrenia genes, which many people even five or six years ago thought would never bear fruit, has turned out to be remarkably fruitful. We now have somewhere between 10 and 20 candidate genes for schizophrenia, the majority of which will probably turn out to be legitimate schizophrenia susceptibility genes. This is a much more successful gene discovery effort than in many other areas of medicine, for example, complex genetics like obesity and diabetes, where much more money has been invested in the search for genes.

You often make the point that the genetics of schizophrenia are the genetics of risk, which is different from those of, say, Huntington’s disease or cystic fibrosis, which you call the genetics of fate. In other words, schizophrenia is polygenic and heterogeneous. Can you explain these terms and their implications for family members of people diagnosed with schizophrenia?

We know from decades of doing family studies, adoption studies, twin studies and now genetic studies that there is nothing you inherit that “gives you” schizophrenia. It’s not like cystic fibrosis of Huntington’s disease or other traditional metabolic inherited disorders, where the gene determines whether you get the disease or not. We’re not dealing with conditions that have a very discrete biology, we are dealing with conditions that are about interacting risk factors, both genes and environment. So, we can’t use genetics to predict with great certainty who will develop disease. We can use genetics to characterize someone’s risk in the same way we measure cholesterol to identify risk for heart disease. You can have very high cholesterol and never have heart disease. You can smoke and never get lung cancer. It’s all these things. This is the notion of susceptibility genes. And this is the story in most common medical illnesses, where multiple genes interact to produce a common phenotype.

Heterogeneity means that no two people, unless they’re identical twins, and no population of people with a common phenotype will have exactly the same genetic background factors. The notion of polygenic means that there’s no one gene that all by itself accounts for the disorder.

How do you feel that the study of the genetics of schizophrenia might impact clinical care over the next decade?

I think it will have enormous impact on many levels. First of all, doctors, as caregivers, now have, for the first time, a sophisticated and informed perspective on the basic causal mechanisms of disease. [In the past], doctors dealing with very difficult and challenging clinical problems [treated them] in the context of essentially a mythology, which is part of the past understanding of what mental disorders was. We [will be] helped considerably by having all of that clinical work take place in the context of a firmer understanding of some basic causal mechanisms. That’s one impact.

Secondly, [genetics] reduces the stigma of the disease. Because the more we elucidate [explain] what the real biological pathways to illness are, the less innappropriate talk there will be about somebody “being a bad actor”, “lazy”, “not picking themselves up by their bootstraps”, etc…also, the more this is like other common medical disorders, where multiple genes interact with the environment to determine one’s likelihood of manifesting illness, the more it looks like [schizophrenia] is in the same realm as diabetes, obesity or heart disease.

Lastly, the genes will allow us to design model systems, in animals and in cell models, that will identify unsuspected, currently unknown targets for the development of new treatments. This will come, in time.

Dr. Weinburger, you are someone trained in psychiatry and neurology, who has studied mental illness for over 25 years, currently at the Federal level. Today you stated that this point in history offers something unique: the ability of genetics to greatly inform all branches of medicine, in particular psychiatry. Can you say a little bit more about this time in scientific history?

This is the first time we’ve had any objective hints to the causes of mental illness. All of our searches prior to this era, which is the era of gene discovery, has been research based on the phenomenology or the epidemiology of mental illness – that is, characterizing risk factors in populations and doing a variety of different assays of increasing sophistication on what’s different about someone who has schizophrenia versus someone who does not have the disorder. The problem with phenomenology (and even epidemiology) is that it doesn’t identify mechanisms, and so that makes it very difficult to know cause from effect. Genes are the first objective clues to the actual mechanisms of disease and basic causes. This is to me is a huge sea change, as are the tools we now have to approach the understanding of mental illness.

Explainatory note on what Dr. Weinberger is saying, and what "phenomenology" is:

In context, he’s referring to the fact that all psychiatric disorders, at present, are defined by clinical phenomena, e.g. observable signs and subjective symptoms. This is the reason that the DSM-IV has a “Chinese Menu” approach: 5 out of 9 of this or that symptom, etc. This is a first step and necessary for consistency and epidemiology, but indicates that we know relatively little about precise biological method of causality of schizophrenia.

To illustrate, 100 years ago one could have defined lung cancer phenomenologically:

“Lung Cancer is defined by the presence of pleuritic chest pain (pain with chest movement) and at least 2 of the following 4 criteria:

-bloody sputum, at least once per month
-chronic cough of at least 3 months duration
-history of daily cigarette smoking for at least 10 years
-any unintended weight loss of 5% or more over the past 6 months
-In addition, the patient must have no signs of an acute infection (fever, sweats, chills) and these symptoms must not be better explained by another illness.”

This would have been “pretty good” at catching “real” lung cancer, but not perfect. Ever since surgery and pathology have gotten good enough, we now define it by its structure, microscopic anatomy and genetics – and the margin of error is now very small. Currently, our technology does not allow us to better define brain disorders (until after death), although we are starting to be able to do so with certain types of dementia.

Dr. Weinberger’s point is that genetics has the potential to allow us to sub-divide psychiatric disease into reliable categories based upon objective testing. This helps us treat the underlying cause of disorders that often have the same behavioral output for different reasons – just like the lung cancer diagnosis we gave above will catch a lot of people who don’t actually have cancerous cells.

Thank You.

It was my pleasure.

More information
The Causes and Prevention of Schizophrenia


Great ideas on the genetics of schizophrenia. Because of its polygenic and heterogenous nature, the genetics of schizophrenia will hardly be clearly understood.

What proteins or neurotransmitter abnormality is inherited that puts the individual at risk of schizophrenia and what is the inheritance pattern? Are there any hope for development of genetic markers in the near future?

This no doubt will help identify subjects who are prone and therefore need to make environmental adjustments to minimise risk of developing the disorder. Are these genes turned on at a particular age for some people?

But one of the confounding factors in the understanding of the genetics of schizophrenia is whethr to begin with is schizophrenia one disorder or a spectrum of mental disorders with similar psychopathological profile, and posssibly similar aetiological basis.

Posted by: Solomon Agbahowe at October 11, 2006 12:17 AM

I agree with this completely, thanks for the post.

Posted by: John at June 27, 2007 07:47 AM

Post a comment

Please enter this code to enable your comment -
Remember Me?
(you may use HTML tags for style)
* indicates required