Samuel Barondes [12.2.03]

Most of the psychiatric drugs we use today are refinements of drugs whose value for mental disorders was discovered by accident decades ago. Now we can look forward to a more rational way to design psychiatric drugs. It will be guided by the identification of the gene variants that predispose certain people to particular mental disorders such as schizophrenia or severe depression.



Psychiatrist Samuel Barondes, M.D. is interested in the ways that chemicals influence mental processes. "Despite decades of tinkering," he notes, "the drugs we presently use still have serious limitations. First, they don't always work. Second, they still have many undesirable side effects. Instead of continuing to invest in more minor improvements, pharmaceutical companies are becoming interested in a new approach to psychiatric drug development." In this discusion he traces the accidental discovery of LSD by Albert Hoffman in 1943 as a contribution to a milieu that favored the discovery of many psychiatric medications.

For example, he notes that "in the course of just a few years there were these two discoveries of extremely valuable psychiatric drugs that radically changed the practice of psychiatry. Before the discovery of chlorpromazine and imipramine disorders like schizophrenia and major depression were usually dealt with by talking, exhortation, and hospitalization—and with limited success. With these new drugs many patients had remarkable improvements."

There are new approaches that take advantage of the fact that there are genetic vulnerabilities to mental disorders. "The hot new technologies that psychiatric scientists are now using," he says, "include not only genetics but also brain imaging...It will be possible to correlate knowledge about genetic variation with knowledge about how specific brains operate in specific circumstances, as looked at with various kinds of functional magnetic resonance imaging. Right now our ideas about mental disorders are mainly based on interviews, questionnaires, and observations of behavior. Being able to look at what's going on inside the human brain, once considered to be an inscrutable black box, is turning out to be quite informative."


SAMUEL H. BARONDES is the Jeanne and Sanford Robertson Professor and director of the Center for Neurobiology and Psychiatry at the University of California, San Francisco. He is past president of the McKnight Endowment Fund for Neuroscience and recently served as chair of the Board of Scientific Counselors of the National Institute of Mental Health.

Barondes is interested in psychiatric genetics and psychopharmacology. He is the author of Molecules and Mental Illness; Mood Genes: Hunting for the Origins of Mania and Depression; and Better Than Prozac: Creating the Next Generation of Psychiatric Drugs. 

Samuel Barondes' Edge Bio Page

The idea that animals can be used to study mental illness strikes many people as strange, because human behavior seems unique. After all, only humans use language for introspection and long-range planning, and it is just these functions that are disturbed in many psychiatric disorders. Nevertheless, we have enough in common with other animals to make them very useful for studies that can’t be done with patients.

Of the animals used for this purpose, apes and monkeys have been favorites because they are our closest relatives. Dogs, too, have obvious human qualities. They may even display patterns of maladaptive behavior that resemble those in DSM-IV. For example, Karen Overall, a professor in the School of Veterinary Medicine at the University of Pennsylvania, has been studying a dog version of obsessive-compulsive disorder (OCD), which is fairly common in certain breeds. Like people with OCD who each have their particular patterns of symptoms, individual dogs with canine OCD also have distinctive main symptoms such as tail-chasing or compulsive licking of their paws. Like humans with OCD, the dogs tend to perform their rituals in private. And, like human OCD, the canine version responds to drugs such as clomipramine and Prozac. Because of these many similarities, Overall’s dogs may provide information about the human disorder—an aim that has already been achieved for another canine behavioral disorder that I will turn to shortly.

But despite their value for certain types of studies, primates and dogs are not ideal experimental animals. Their main shortcoming is that they are costly to raise and maintain, which makes them impractical for the many experiments that require large numbers of subjects. For this reason scientists have been turning to a much less expensive alternative, the laboratory mouse. Although it is more difficult to empathize with these tiny rodents than with a chimpanzee or a golden retriever, we now know that all these mammals share much of our complex brain machinery. What makes mice especially attractive is that their genes are relatively easy to manipulate by traditional breeding methods and by the new techniques of genetic engineering. Both experimental approaches have been successfully employed to make special strains of mice that are being used to study mental disorders and to develop new psychiatric drugs.


(SAMUEL BARONDES): I'm interested in the ways that chemicals influence mental processes. Most of these chemicals are brain proteins whose structure and expression are controlled by genes. But some of them are simple chemicals such as serotonin and norepinephrine, which transmit signals in the brain, signals that are important for experiencing emotions. These simple chemicals are also of great interest because their actions in the brain are influenced by widely used psychiatric drugs, such as Prozac. It was, in fact, the accidental discovery of several mind-altering drugs in the middle of the 20th century that drew me into research on brain functions and mental illness.

The most legendary of these discoveries was made in 1943 by Albert Hoffman, a chemist who worked at Sandoz, a Swiss pharmaceutical company. While making a variant of a chemical that causes uterine contractions he noticed that his mind was playing tricks on him. Suspecting he had inadvertently swallowed a bit of the drug, Hoffman took a tiny measured dose. He was astounded to find that the drug changed his perception of common objects, intensifying their colors and altering their shapes. The drug he created is LSD.

In those days there were not many limitations on the pharmaceutical industry. So Sandoz distributed LSD to a variety of researchers in the hope that they would find a medical use for it. Attempts were made to use LSD as an aid to psychotherapy, but this application never caught on. And it did not take long before some of the drug that was provided to researchers began to be diverted for recreational use. As LSD's popularity grew more and more people began to realize that their minds could be controlled by minuscule amounts of certain chemicals. If anyone had any doubt about this, a dab of LSD would persuade them otherwise.

Although LSD did not prove medically useful, it contributed to a milieu that favored the discovery of many psychiatric medications. The first of these was synthesized in 1950 by Paul Charpentier, a chemist who worked for Rhone Poulenc, a French drug company. Charpentier had already made his mark by creating promethazine, a drug that blocks the action of histamine. Promethazine, which was marketed as Phenergan, is one of the early antihistamines that are still used to treat symptoms of allergies and colds. Charpentier's subsequent discovery of a revolutionary psychiatric medication was stimulated by the observation that antihistamines make you sleepy, which suggested that histamine plays a role in the function of the brain.

To people who use antihistamines to relieve allergic symptoms sleepiness is an undesirable side effect. But Henri Laborit, a French naval surgeon, turned this to his advantage by using promethazine as an adjunct to anesthesia. Once he discovered this application for promethazine he contacted the Rhone Poulenc Company. They asked Charpentier to make derivatives that are more effective in inducing sleep.

In 1950 Charpentier made a new derivative that seemed to fit the bill. Among the doctors who experimented with it were two French psychiatrists, Jean Delay and Pierre Deniker. They tested it on agitated patients in a psychiatric hospital and found that it would calm them down. To their great surprise they also found that some patients with schizophrenia who took the drug for several weeks stopped hearing nonexistent voices and stopped worrying about nonexistent plots against them. With continuous treatment many patients who had been hospitalized for years got progressively better. Promethazine, the antihistamine from which this new drug, chlorpromazine, was derived, had none of these antipsychotic effects. So there was some additional property of this new antihistamine that made it very different from the old ones. It not only made people sleepy, it also changed the thoughts of people with schizophrenia. Their paranoid delusions often disappeared, and their ability to relate to other people improved. This was an extraordinary discovery. By 1955, just a few years after the discovery of these properties of chlorpromazine, it became a blockbuster drug with the trade name Thorazine.

Stimulated by the great success of chlorpromazine, other drug companies began searching for drugs for schizophrenia. Among them was Geigy, a Swiss pharmaceutical company that was also making antihistamines, including one that looked very much like chlorpromazine. But when Geigy distributed their drug to psychiatrists they got disappointing results—their antihistamine was quite useless for schizophrenia.

Luckily for Geigy Roland Kuhn, one of the psychiatrists who got a supply of this drug, was interested in depression, and he gave it to people with severe depression. To his great surprise the depressions started lifting after a few weeks of treatment. By 1958 Geigy's antihistamine, called imipramine, or Tofranil, had become another blockbuster. 

So, in the course of just a few years there were these two discoveries of extremely valuable psychiatric drugs that radically changed the practice of psychiatry. Before the discovery of chlorpromazine and imipramine disorders like schizophrenia and major depression were usually dealt with by talking, exhortation, and hospitalization—and with limited success. With these new drugs many patients had remarkable improvements.

Once these drugs were discovered a few scientists began to examine their effects on brain chemistry. Most of this research was done at the National Institutes of Health, rather than in pharmaceutical companies. Within the course of the next ten years it was discovered that both chlorpromazine and imipramine influence the actions of brain chemicals, called neurotransmitters, which transmit signals between brain cells. Chlorpromazine was found to block receptors for a neurotransmitter called dopamine, and imipramine was found to augment the actions of two brain chemicals, norepinephrine and serotonin. Both drugs have many other effects on brain chemistry, but their modifications of neurotransmission are believed to be responsible for their therapeutic actions.

Once these brain effects were discovered pharmaceutical companies set out to find new drugs that retain the therapeutic effects of chlorpromazine or imipramine but that lack their undesirable properties. Over the years they created a number of new drugs. For example, we now have a variety of new drugs for schizophrenia including Risperdal, Zyprexa, Seroquel, Geodon, and Abilify. All block dopamine receptors in the brain, a property they share with chlorpromazine. But each has other properties that make it a better drug. We also have a variety of new drugs that share certain properties of imipramine. Among them are Prozac, Zoloft, Paxil, Luvox, Celexa, Lexapro and Effexor. All are descendants of imipramine; and all are free of some of imipramine's undesirable side effects.

But, despite decades of tinkering, the drugs we presently use still have serious limitations. First, they don't always work. Second, they still have many undesirable side effects. Instead of continuing to invest in more minor improvements, pharmaceutical companies are becoming interested in a new approach to psychiatric drug development.

The new approach takes advantage of the fact that there are genetic vulnerabilities to mental disorders. For example, if you have a mother or a father with schizophrenia, your chances of developing schizophrenia are about nine times as great as that of other people. One out of every hundred people in America, or in the world, is schizophrenic, but nine out of a hundred children of a parent with schizophrenia develop this disorder, usually by early adulthood. Likewise, for people who develop severe depression early in life—before approximately age 20 to 25—the risk to their children is about four or five times as great as that to the rest of the population. This and other information point to a genetic vulnerability to these mental disorders. This doesn't mean that if you inherit the gene variants or the combination of gene variants that increase this vulnerability, you will inexorably develop the condition. Nonetheless, now that the human genome has been sequenced, and now that the variations of human genes are being tabulated, the stage has been set for the discovery of the gene variants that are responsible for such vulnerabilities. Finding these gene variants will lead to the identification of the brain circuits whose functions they affect. This, in turn, will point the way to the development of new medications that help to normalize these brain functions. Hopefully, drugs designed in this way will be better than the ones we now have.

Of course, all this remains to be seen. It's one thing to know that a biological variation exists, but it's not easy to figure out a way to alter the consequences of this variation with a drug. Furthermore, drugs designed in this way are just as prone to side effects as those that were stumbled upon in the past. Nevertheless, this new approach should eventually provide us with highly effective new medications for depression, schizophrenia, and the other prevalent mental disorders that plague mankind.


My interest in psychiatry didn't begin with an interest in brain biology. When I went to Columbia College in the 1950s I was mainly interested in studying human behavior. In the Columbia of that era the main players were Sigmund Freud and B.F. Skinner. Freud's influence was still very great, and his Civilization and its Discontents was required reading for freshmen. But in the psychology department of Columbia College Skinner's behaviorism ruled, whereas Freud was viewed with great suspicion. I became captivated by both types of ideas about behavior and the mind, but it seemed to me that Skinner's quantitative behavioral experiments were more likely to be productive than Freud's clinical observations and theories. There was this great battle between them. Skinner said all that Freudian stuff was baloney because it was just made up, and one could not test it experimentally. And the psychoanalysts said in response, "Look at all the interesting things we talk about; all you talk about is lever-pressing in rats." I grew up in that milieu. Those were the interesting issues in the mid-'50s.

I liked Skinner's work very much, but I also really liked chemistry and biology, and I had the sense that maybe these hard sciences could be applied to psychological problems. This led me to medical school, but I was soon disappointed because the psychiatrists at Columbia Medical School were in that era very dogmatically psychoanalytic, and not experimental, and not open to discussion. They were very orthodox in that period. So I transiently changed directions and turned to endocrinology. I became interested in hormones because hormones affect the brain, and I wound up doing some work in internal medicine. Then I did a postdoctoral fellowship at the National Institutes of Health. This was in the early '60s, when the National Institutes of Health was absolutely in its golden age. Through a series of lucky circumstances, I arrived as an endocrinologist and ran into a man named Gordon Tomkins, who was also an endocrinologist and one of these natural teachers and avuncular people who would gather young people around him. He became my mentor, and I remember he took me into his office and said, "You know, endocrinology is really molecular biology. It's all genes. What hormones do is regulate the function of genes." This was in 1960 or 1961 and his vision has been completely verified. This was before we had good tools to study them. He said, "If you're interested in this stuff, and you can work on the mind too, just go and study molecular biology."

Molecular biology was completely new. I didn't even know what it was. Remember, the Watson/Crick double helix was discovered in 1953, and this was 1960—not that long afterwards, and the double helix was not generally appreciated—and very few people were available to teach me how to become a molecular biologist. Fortunately Gordon had a young person named Marshall Nirenberg in his group, and Gordon suggested him. But Marshall was unknown at the time, and I said, "I want to work with you, Gordon." He said, "No, you can't work with me, because you don't know nothin', so go work with Marshall. He needs somebody in his lab, so work with him." Three weeks after I joined Marshall, he discovered that polyuridylic acid, a polynucletide which is a string of uridines, instructs the protein-synthesizing machinery to make an unusual protein made up of a string of phenylalanines. This suggested that the sequence u-u-u—uridine, uridine, uridine—codes for the amino acid phenylalanine, and this was the beginning of the deciphering of the genetic code. It is the key to understanding how the language of nucleic acids—the language of the genes—is translated into the language of proteins which control the functions of living things. As a complete novice of a doc, a knee-tapping, stethoscope-carrying doc, I suddenly found myself working on one of the great scientific problems of the time, the genetic code. About 6 or 7 years later Nirenberg won the Nobel Prize for the genetic code. And if you open up Albert's cell biology book, Molecular Biology of the Cell, the inside cover shows the genetic code in which u-u-u encodes phenylalanine and other strings of three nucleotides encode the other amino acid components of proteins. The genetic code has the same importance in molecular biology that the Periodic Table of the Elements has in chemistry. If you open any chemistry book there's the table of the elements; in molecular biology or biology books it's the genetic code.

All of a sudden I found myself—just completely by good luck—immersed in molecular biology and in the company of some really famous people. Soon Crick came to visit, and Watson came to visit. Gordon also encouraged me to follow my interest in the brain and in psychiatry, so I did a residency in psychiatry, knowing that I would not have to be a psychoanalyst because I now knew how to do science. I had been transformed into a young molecular biologist, and I could try to bring molecular biological approaches to research on mental illness. Through a series of accidental circumstances my career was set, and ever since then I've been trying to use molecular biological tools to solve problems that are relevant to psychiatry.

I am very interested in studies of the genetics of mental disorders because they take advantage of the accumulation of vast amounts of knowledge about human genetic structure and genetic diversity. The next simple-minded thing to do is to identify gene variants that influence particular behavioral propensities, and that's going to be doable in the next 5, 10, or 20 years. The basic sequence of the human genome is known. It's known that there are a small number of very common gene variants. What we need to do is correlate the various gene variants with various behavioral propensities. And as the ways of crunching genetic data are improved, as it becomes cheaper and cheaper to take DNA samples from large numbers of people and look at all the variants in each individual person, and as computers are available to integrate all that data, we're going to learn a lot about the genetic propensities towards different kinds of human behaviors. That's a certainty; that's definitely going to happen.

A lot of it comes down to economics. When sequencing DNA samples from large numbers of people becomes affordable—which will probably happen fairly soon, for research purposes at least—we will be in a position to learn a lot about gene combinations and the propensities to certain kinds of behavioral traits. Or we will find that we can't figure it out. That may happen too. It may turn out that although certain propensities are heritable they're so complicated—there are so many genes involved—that we can't say a great deal about it. I suspect that that's not always going to be the case. I suspect that there are going to be some important gene variants that are going to have significant functions in sorting out general directions of personality. It's an experimental question that will be settled one way or the other, and it's certainly the way to go. The technology is available, and becoming available, and its importance is tremendous. It's going to be important for understanding ourselves, and it's also going to be important clinically for helping people with various kinds of mental distress. There's going to be this huge repository of genetic variations and attempts to correlate them with certain behavioral propensities. And although it's going to be really complicated, and a lot of the stuff is going to be undecipherable because the changes in propensities might be one percent or two percent—stuff that's going to be hard to pick out of the noise or hard to say is really important—I suspect that there are going to be some genetic variations that have substantial effects on the risk of certain mental disorders.

The reason I have hope is the Alzheimer's story. You know Alzheimer's disease, the common kind that happens in old people—not the rare early-onset form, which is a Mendelian disease that happens to people before 50, but the very common older-onset form which usually begins to strike at the age of 70 or 80. The older-onset form is influenced, very importantly, by variants of one gene: the Apo-E gene. The Apo-E gene is a gene that encodes a protein that transports lipids in the blood. There are three variants that are each very common in the human genome—Apo-E 2, Apo-E 3, Apo-E 4. What's become clear is that if you inherit two Apo-E 4s, one from mom and one from dad, then your risk for Alzheimer's Disease by the age of 70 is about 50%. You have a huge increase in risk for this terrible, dementing illness, this change in your personality and cognitive ability. It's not the whole story—there are people with two Apo-E 4's who live to be 90 and don't get Alzheimer's Disease—but it's an example of a common gene variant, which if you happen to inherit it, especially in two copies, significantly increases your risk of a huge mental change in the progress of life.

That may turn out to be an unusual discovery; that is, it may turn out that none of the genes that control other important psychological oddities—like becoming schizophrenic or manic depressive—have the same propensity to do this that the Apo-E 4 gene does for Alzheimer's Disease. Their power, the percentage variance that they are responsible for, may be significantly less, but there are going to be some gene variants that are going to turn out to be really important in increasing the risk of developing such mental disorders, and the hope is that we're going to learn a lot from them—among other things about how to make new drugs. Once you find a gene you're going to be able to find out which enzyme, structural protein, or regulatory protein it codes for, which in turn gets you into a functional biological pathway, which in turn can help you design a drug.

The story of Alzheimer's Disease is the story of showing how one very common gene variant has powerful effects on one form of behavioral problem, albeit an odd one, one that progressively occurs in old age, but one that we're all very interested in. Apo-E 4s is not a rare, weird mutation. It's a very important variant of a gene that everybody has. This gives me hope that there will turn out to be other common gene variants that make people prone to things like depression. There may be really powerful ones that account for a significant part of the variance. Knowing about those will really help us a lot in diagnosis, and also in designing new medication. It remains to be seen how well this works out, but the neat thing about the genome and the gene variant stuff that's being explored now is that there's going to be this pile of information and it's going to be analyzable, because computer technology is such that it can be parsed out. We're going to learn a lot. How useful it's going to be remains to be seen, but let us hope that it will be very useful.

The hot new technologies that psychiatric scientists are now using include not only genetics but also brain imaging. Brain imaging has brought another dimension to studies of human brain functions because you can, in real time, look at brain regions that are active in certain kinds of mental processes, and can look at differences in different people—that is, people with different conditions or propensities—to see in real time how their brains might be operating. There are going to be opportunities to use knowledge about specific gene variants that can be tied to this imaging, so that one can look at more than just measures like behavior that you assess with conversation or questionnaires. It will be possible to correlate knowledge about genetic variation with knowledge about how specific brains operate in specific circumstances, as looked at with various kinds of functional magnetic resonance imaging. Right now our ideas about mental disorders are mainly based on interviews, questionnaires, and observations of behavior. Being able to look at what's going on inside the human brain, once considered to be an inscrutable black box, is turning out to be quite informative.