SCIENCE

Creative animal models screen drugs

Psychologists are continually developing and refining animal behavioral models to test drug efficacy

By Bruce Beans

Using ingenious animal protocols that probe the subtle behavioral and psychological effects of drugs on animals, psychologists are playing an increasingly important role in pharmaceutical firms' early-stage drug development.

The creativity of these psychologist-designed behavioral animal models lies in the fact that even though they don't always exactly mimic a human condition--such as anxiety or schizo phrenia--they are practical and highly repeatable scenarios that can reliably predict the effects drugs might have on humans.

"These procedures have been used in countless laboratories and, for certain classes of drugs, have high predictive validity," says Nancy Ator, PhD, past president of APA's Div. 28 (Psychopharmacology and Substance Abuse) and professor of behavioral biology at Johns Hopkins University School of Medicine. "There's a high correlation between these screening procedures and the likelihood, for example, that a compound will effectively treat anxiety in humans."

Psychologists have used animal models to guide preclinical drug testing since 1954 when the FDA approved chlorpromazine hydrochloride (Thorazine) for schizophrenia. That development, and the subsequent approvals of chlordiazepoxide (Librium) in 1960 and diazepam (Valium) in 1963, led many pharmaceutical firms to establish preclinical behavioral pharmacology laboratories headed by psychologists.

Today, the pharmaceutical industry is increasingly depending on psychologists' animal model studies as essential early screens for the most promising compounds.

"We're using animal models more in part because they have become so sophisticated," says Robert Mansbach, PhD, a senior research investigator with Pfizer, Inc.

These studies, he says, are key to industry-wide efforts to boost the number of drugs that earn FDA approval. And, backed by the support of pharmaceutical companies, psychologists are continually refining and developing new animal behavioral models.

Future models

Among those new models is one developed two years ago by Kenneth J. Sufka, PhD, in the University of Mississippi's Department of Psychology. Sufka sought to develop a model that used chicks rather than more expensive rats to cost-effectively screen prospective anti-anxiety drugs. If successful, the chick model would measure the effect of the compounds on the central nervous system's benzodiazepine receptors, which moderate human anxiety.

In the research, Sufka and his colleagues separated the chicks from their social companions. Under such stress, the chicks increased their distress calls. When the chicks were given Librium, a proven anti-anxiety compound known to affect the benzodiazepine receptors, they calmed down.

To prove for sure that separating chicks from their social companions was indeed a model of anxiety, Sufka needed to be able to reverse or block the effects of the Librium.

For that, he needed flumazenil, a controlled substance used to reverse sedation or general anesthesia, as well as to treat patients who are suspected of having overdosed on anti-anxiety drugs, such as Librium and Valium. Intrigued by his proposal, the drug's manufacturer, Roche Pharmaceuticals, donated a supply of flumazenil to administer to the chicks.

The drug did, indeed, reverse the effects of the Librium on stressed chicks--an effect that implicates benzodiazepine receptors in modulating chick distress calls. As a result, says Sufka, whose study appeared in the May issue of Experimental and Clinical Psychopharmacology (Vol. 7, No. 2, p. 83-89), "We think this particular animal paradigm might be useful for screening anti-anxiety drugs."

Already, one pharmaceutical company is interested in having the University of Mississippi's National Center for the Development of Natural Products use Sufka's chick model to screen the company's natural product compounds for anti-anxiety effects.

The next frontier

While it is still primarily the purview of molecular biologists and psychiatrists, the field of genetics also holds great promise for such psychological preclinical research, believes James Barrett, PhD, vice president of neuroscience research at Wyeth-Ayerst Research in Princeton, N.J.

Across the country, research laboratories are creating "knock-out mice"--the genetically mutated mice that are bred with specific genes inactivated. Among the knocked-out genes being isolated for pharmaceutical research are those thought to affect drug addiction and serotonin receptors.

Serotonin, for example, acts on at least 14 different receptor types in the brain, all of which are likely to be affected by drugs such as Prozac.

"It would be useful to know which of these receptors is important for the beneficial effects of such drugs," says Laurence H. Tecott, MD, PhD, an associate professor of psychiatry at the University of California-San Francisco.

To do that, Tecott and his colleagues have genetically knocked out specific receptors. For example, some of his mice lack a serotonin-receptor subtype that, when stimulated, is thought to increase anxiety levels and suppress feeding behavior. Lacking such a receptor, these mice overeat throughout their lives and show a decreased sensitivity to the appetite-suppressant drug fenfluramine.

Tecott also has knocked out another serotonin receptor called serotonin 1A, which is thought to suppress anxiety when it is stimulated.

In one behavioral experiment, these mice are placed on an elevated runway that is divided into quadrants--two sections enclosed by walls, two open. The researchers find that the mice spend much more time in the relative security of the walled sections than normal mice: They apparently act more anxiously because they lack a gene that suppresses anxiety.

A decade ago, many researchers assumed genetics would eventually render behavioral animal research obsolete because it would be able to link specific genes to specific behaviors, says Jackie Hunter, PhD, group director of neurobehavioral research for SmithKline Beecham in England.

But Tecott still relies upon proven animal behavioral models to shed light on the effects of his genetic manipulations.

Also, notes Hunter, it is now thought that many human disorders are not controlled by single genes.

"Sometimes, if you knock a gene out in an animal, you get a very clear effect," she says. "But, more often than not, it is very subtle. It takes behavioral pharmacologists or psychopharmacologists to tease out what the effect actually is."

Drug-abuse liability

In addition to researching drug therapies, psychologists have long conducted preclinical studies on drugs that have the potential to combat addictions to nicotine and illegal substances. With funding from pharmaceutical companies, they are also applying their knowledge of addiction to gauge whether a new drug treatment could be addictive--what is known as the drug's abuse liability.

William L. Woolverton, PhD, is one of several researchers who employ well-known addiction models. Woolverton, professor and director of the Division of Neurobiology and Behavior Research at the University of Mississippi Medical Center, works primarily with monkeys, but also with rats and pigeons.

In one model that he uses to test both anti-addiction drugs and the abuse liability of drugs, Rhesus monkeys press levers to self-administer drugs through IV lines. Typically, three to four monkeys that readily take a drug of abuse, such as cocaine, are involved in a three- to six-month test of each prospective drug. The monkeys often act as their own controls having alternate access to the drug being studied and to a placebo saline solution.

"It's an outstanding animal model of drug abuse in humans," Woolverton says. "Monkeys tend to self-inject the same drugs humans abuse with similar rates and patterns of tolerance and dependence."

Woolverton also conducts tests that enable him to determine how a drug might "feel" to humans. For example, to detect cocaine-like effects, he injects a monkey with cocaine, then gives it food if it pushes a right-handed lever. He then reverses the procedure by injecting the animal with a saline solution and rewarding it with food only if it pushes a left-handed lever. Subsequently injected with a drug under investigation, the monkey would press the right-handed lever if the drug felt similar to cocaine.

"Self-administration tells you whether an animal will take a drug," says Woolverton. "The drug discrimination test tells you what it feels like."

Belinda Hayes, PhD, an FDA pharmacologist, says both the preclinical and clinical abuse liability studies are routinely used by the FDA in determining the appropriate level of control for a new drug under development, or for other substances under the Controlled Substances Act.



Read our privacy statement and Terms of Use

Cover Page for this Issue

PsychNET®
© 1999 American Psychological Association

APA Home Page . Search . Site Map