New Research on Antidepressants
by Maia Szalavitz
Posted December 22, 2000 · Issue 93
Recent research into the molecular mechanisms of antidepressants offers new treatments for depression and a better understanding of this common mental illness.
For decades, neuroscientists have sought to understand how antidepressants work and why so many drugs with seemingly varied, even opposing, mechanisms of action can be used to stave off lingering black moods with some degree of success. Though it commonly is believed that raising serotonin levels brings relief, experts have long known that this explanation is contradicted by data and is far too simple to be correct.
Recently, a new theory about antidepressant action has been gaining attention and support. This set of hypotheses may help resolve questions not only about how antidepressants work, but about what successful talk therapies do, and what causes some types of depression.
Eric Nestler, chairman of the Department of Psychiatry at the University of Texas Southwestern Medical Center at Dallas, says, "The major challenge is that antidepressants are a very diverse group of chemicals and their protein targets vary. Some act on serotonin, some work on the norepinephrine transporter, others bind to chemicals we haven't even identified. When you give them to animals, you can find some common changes, but the real challenge is to relate specific molecular adaptations to antidepressant action.
"Having said that," Nestler continues, "There do seem to be several common changes caused by all antidepressants, with no exceptions that I know of." One such change is an increased presence of a chemical called cyclic AMP response element binding protein (CREB) in the brain's hippocampus.
The hippocampus is known for processing and sorting memories for storage - without it, new long-term memories cannot be created. In a famous and horrific case of severe damage to the hippocampus, a patient known as H.M. was left permanently stuck in a moment of time by an operation aimed at curing his epilepsy that, instead, destroyed this area on both sides of his brain. His long-term memories of what had happened before the surgery were intact, but when H.M. met new people, even when they visited him repeatedly, he would forget that he'd ever known them five minutes after they'd left.
One physician performed a somewhat nasty experiment on H.M. to determine whether he could form new emotional memories. The doctor greeted him, shook hands, and poked him with a pin while doing so. When this doctor returned later, H.M. had no conscious memory of their previous meeting but, while he would shake hands with other physicians, he refused to do so with this one. It seems that proper functioning of the hippocampus is needed to store memories and integrate them with the emotions they evoke.
CREB is one of a number of "response elements" that stimulate the activation of particular genes. In the hippocampus, CREB stimulates production of several proteins, one of which is known as BDNF (brain-derived neurotrophic factor). BDNF was discovered because it plays a role in neural development in young animals, but in the adult hippocampus, it seems to preserve and protect neurons from stress and encourage them to make new connections. "With more BDNF, a neuron is healthier and happier," says Nestler.
Other research has found that stress - particularly early, sustained stress - can kill cells in the hippocampus. The human stress hormone cortisol may be particularly dangerous to these cells. About 50 to 60 percent of depressed people seem to have elevated levels of cortisol and other abnormalities in their stress regulatory system. This finding particularly intrigues researchers, since clinical data show very consistently that childhood stress, such as abuse or the loss of a parent, can increase the risk of depression. Even better for the theory - all antidepressants normalize this system if they are effective, and a return to elevated cortisol levels seems to signal that a formerly depressed person is about to relapse.
What antidepressants may do, then, is help the hippocampus recover from the effects of sustained stress. Because the hippocampus is involved in memory integration, it's not hard to imagine that one consequence of damage to this area would be difficulty coping with painful memories and experiences that recall them. As a result, talk therapy that works might also have its effect here. Both the drugs and the therapy may encourage and nurture these vital cells. "It's a very interesting possibility," says Nestler.
Michael Owens, associate professor of psychiatry and behavioral sciences at the Emory University School of Medicine, has studied the relationships between stress hormones and antidepressants. "Corticotropin-releasing factor (CRF) controls the endocrine response to stress and causes the secretion of cortisol," he says. Blocking CRF, therefore, may be another way to fight depression. The animal research supporting this is strong enough that there are currently, in early clinical trials, about a half dozen CRF-blockers produced by various drug companies.
"We are getting ready to publish a paper looking at whether different antidepressants decrease CRF," Owens says. "These drugs do nothing in normal rats, but in rats that are stressed enough to show changes in their CRF levels, antidepressants seem to be able to bring them towards normal."
Ronald Duman, professor of psychiatry and pharmacology at Yale University School of Medicine, points out other studies that support the hypothesis that depression can result from an overload of stress hormones in the hippocampus. "There is brain imaging data showing a decrease in the size of the hippocampus in depressed patients," he says and adds that there will soon be data on whether successful antidepressant treatment restores it to normal. Also, infusions of BDNF into rats' brains show a potent antidepressant effect.
This effect, like that of other antidepressants, however, takes a while to appear. One important problem with the hypothesis that low levels of serotonin or other neurotransmitters cause depression is that antidepressants act on these chemicals within hours, but antidepressant effects are seen in weeks. If the growth of neurons or their repair is what is needed for recovery, this would take time and would explain why the drugs take so long to kick in.
Serotonin hasn't been completely taken out of the picture, however. In fact, converging evidence shows that all antidepressants do affect serotonergic neurotransmission by making it more efficient. Some do this by raising levels of serotonin itself, directly or indirectly via feedback from other transmitter systems. Others may lower levels of serotonin, but increase the sensitivity of serotonin receptors or increase their numbers. "The consensus is that antidepressants augment serotonergic transmission," says Owens. And because the serotonin system is intimately connected with the system of receptors for stress hormones, this could be one link in the chain of events that restores the stress system to proper balance for those people whose depression is linked to an imbalance.
The question of whether there are different types of depression that require different types of medications remains open. David Healy, director of the North Wales Department of Psychological Medicine at the University of Wales College of Medicine has written a book on the history of antidepressants, The Antidepressant Era (Harvard University Press, 1997).
"My view is that there is a wide variety of pills and that they don't all work the same way," he says. "People focus more on what the drugs do in the brain, when we should be looking more at what they do clinically. We tend to answer the question of what these drugs do and how they do it by saying something about serotonin, but that's not what the drug does, it's where it goes. We're much further from an answer about what they do than we like to think."
Healy has traced various theories about antidepressant action, including the theory that low serotonin causes depression and the earlier idea that depression was caused by low levels of other neurotransmitters. Before there was a focus on common mechanisms, he says, physicians were more likely to look at individual factors that could influence the problem.
Healy recently studied the impact of two antidepressants on normal people and found that different personality types reacted in different ways. "Those who were more reward-dependent responded better to an SSRI, while those who were harm-avoidant responded better to reboxetine," he said.
Since the subjects were normal to begin with, the response seen was not antidepressant, but more of a general reduction in stress and a tendency to be more relaxed and comfortable with others. This change was large enough to be noticed by the subjects' friends and partners. However, those who took a drug that didn't match their personality type actually became more uncomfortable and unhappier.
Healy believes that this type of effect could explain the common finding that, in clinical trials, the difference between the effect of antidepressants and placebos is not as large as with other classes of drugs. "In people whom the drug suited, the response was so strong you couldn't mistake it for placebo," he says, "But if we averaged the responses across groups, they wouldn't be very different than placebo."
According to Healy, the theory of depression as a generic "bug in the brain" that can be corrected by chemicals has left out a lot of individual factors. Part of this may be due to financial interests of drug companies, who could refine their clinical trials to determine which patients benefit most from which drug, but don't because simply proving a drug is better than a placebo allows it to be used in a larger market.
As results from the Human Genome Project come in, however, it may be possible to understand more about the different types of depression and how they relate to what happens in the brain. This will swing the pendulum back from major common mechanisms to how they play out in individual people and could allow a greater precision and refinement in treatment than is currently available. Perhaps then the black dog will have finally met his match.
Maia Szalavitz is a health/science journalist who has written for the New York Times, the Washington Post, Newsday, New York Magazine, Salon, and other major publications.
Frederick H. Carlson
is a professional artist and illustrator whose clients include The Saturday
Evening Post, Baltimore Sun and Pittsburgh Magazine.