Perspectives On Medical Research
Volume 4, 1993
Critique of an Animal Model for Testing the Anti-Anxiety Potential of the Experimental Drug NO-328
Murry J. Cohen
Introduction
Dr. James E. Barrett is attempting to study the anxiolytic (anti-anxiety or tranquilizing) properties of an experimental drug called NO-328 by observing its effects in squirrel monkeys. His research, is being conducted at The Uniformed Services University of the Health Sciences, in cooperation with the Henry M. Jackson Foundation for the Advancement of Military Medicine. The project is entitled Experimental Assessment of NO-328: A GABA-Uptake inhibitor with Potential Anxiolycic Activity.(1) Extramural support was provided by NOVO Industri A/S, to whom "any material stemming from the conduct of these experiments will first be submitted.''
Purpose of the Proposed Experiment
The major anxiolytic drugs currently available are the benzodiazepines, such as diazepam or chlordiazepoxide. These drugs work by chemically altering a specific receptor present on neuronal cell membranes, a reaction that involves the neurotransmitter gama-aminobutyric acid (GABA). Benzodiazepines act immediately and show great efficacy in reducing anxiety. However, they may produce certain undesirable effects, such as sedation, impaired cognition, potentiation (magnification of the effects) of other central nervous system depressants, and development of tolerance, dependence, and addiction.(2)
A newer type of anxiolytic is bupropion, a serotonergic drug that works by increasing the level of the neurotransmitter serotonin in the nervous system. Less is known about this drug's effectiveness than that of the benzodiazepines. Bupropion can require several weeks before taking effect, and the side-effects in some patients can be unpleasant. Furthermore, its efficacy is reduced in many patients previously treated with benzodiazepines. It does not, however, seem to result in tolerance and dependence.
In an attempt to find a useful anxiolytic drug without the disadvantages of the benzodiazepines and serotonergics, Barren is studying drugs with different mechanisms of action. One such drug is NO-328 (R-(-)-l-(4,4-(3-methyl-2-thienyl)-3-piperidine carboxylic acid hydrochloride), whose mechanism of action seems different from that of the two classes of currently available anxiolytics. According to Barrett, this drug is "a potent and selective inhibitor of the uptake of 3H-GABA [and] may represent a new type of anxiolytic, with its primary mechanism of action through the GABA system" rather than the benzodiazepine receptor system.(1) This means that the drug directly inhibits the enzymes that normally break down GABA, causing higher GABA concentrations in the nervous system, without necessarily interacting with the benzodiazepine receptor system. If GABA functions directly as an anxiolytic ligand, without requiring biochemical interaction with other benzodiazepine receptor chemicals, then NO-328 and related drugs might function as anxiolytics through a novel, safer, and more effective mechanism of action, and would represent a contribution to our drug arsenal.
Barrett reports that his experiment "is designed to examine the
specific potential anxiolytic activity of NO-328 in a well-characterized
and widely-accepted primate animal model for the detection of anxiolytic
drugs. This model, which employs the maintenance of a selected response
by periodic presentation of food, [followed by] suppression of that behavior
by punishment, has been used extensively in behavioral pharmacology for
studying anxiolytic drugs and correlates well with clinical efficacy." (sic)
Such reliance on animal experiments to explore the "behavioral pharmacology" of
psychotropic drugs for potential application to human beings is indeed
common.(3) Barrett himself has devoted much of career to such research,
specifically investigation of the squirrel monkey's behavioral responses
to pharmacologic interventions and has added to a large amount of data
on this subject.(4)
Summary of the Proposed Experiment
Barrett's experimental paradigm consists of a commonly used laboratory procedure to test potentially anxiolytic drugs. In his experiment, six squirrel monkeys who have become adapted to a primate chair are to be taught to press a lever for food under both nonpunishment and punishment circumstances. Punishment, delivered by administering an electric shock to the monkeys' tails, has predictably been found to "suppress responding to [lever pressing] levels. . .below those normally occurring."(1)(sic) According to Barrett, "It is the performance under this condition. . . that provides the most meaningful information on whether a drug is or will be a clinically effective anxiolytic compound. All such drugs will increase this punished responding. . ."
After the experimental sessions, the monkeys "are returned to their home living cages in a large colony room with visual and auditory access to other monkeys." The anxiolytic and sedative effects of NO-328 are determined by analyzing the drug's effects on both punished and unpunished behavior. The anxiolytic effect is ascertained by seeing if the drug results in an increase in punished behavior, and the sedative effect is gauged by looking for a decrease in unpunished behavior.
The experimental drug is administered through a tube which, over the course of 1-2 minutes, is inserted into the monkeys' stomachs. Some of the monkeys will be given only placebos; in their case, the stress, anxiety, and apprehension brought on by the experimental procedure will definitely be unalleviated: "The compound under investigation may have stress alleviating actions, but monkeys will not always be administered this compound." Barrett assigns the "level of pain or distress to be imposed" a value of "three" on a scale of one to five.
Critique of the Proposed Experiment
Considering the difficulties encountered with the effective but problematic benzodiazepines and the less-than-optimally-effective serotonergic anxiolytics, the attempt to identify an improved anxiolytic drug is a worthwhile endeavor. The subject of Barrett's experiments is not trivial. His chosen methodology, however, warrants serious criticism into two categories: criticism of the animal model approach per se and criticism of other aspects of the experiment.
Criticism of the Animal Model Approach
Animal models have long been used to study human psychiatric illnesses,(5) despite numerous objections to this approach.(6-13) Promoters of animal experimentation have convinced much of the public that psychopharmacologic breakthroughs have derived directly from these experiments.(14) However, Reines' comprehensive historical review revealed that "The discoverers of modern psychopharmacology were clinical investigators who noted mood-altering side-effects of drugs used in clinical practice."(11)
Just as basic science investigators and animal experimenters have inappropriately claimed credit for many other medical breakthroughs,(15) they have claimed credit for the development of psychotropic drugs. A publication of the Foundation for Biomedical Research states that "medications to treat mental illness . . . required animal research in their development."(16) A former president of the American Psychiatric Association has written that "animal research has provided medicines for use against schizophrenia, depression, panic disorder, obsessive-compulsive disorder, and manic-depressive illness."(17) Such statements, however, "rewrite medical history."(12)
Let us examine the actual circumstances that led to the discovery of different classes of psychotropic drugs. Psychiatrist and neuroscientist Ross Baldessarini notes with respect to the phenothiazines: "Like most of the useful psychotropic agents, the effects of chlorpromazine were discovered by serendipitous clinical observation."(18) This fact is completely ignored by J.K. Krasney, who states that the development of chiorpromazine and its tranquilizing derivatives resulted from studies with rats, rabbits, and monkeys.(19)
Serendipity in Discovery of Benzodiazepines details Dr. Leo Sternbach's in vitro discovery of the first benzodiazepine anxiolytic, chlordiazepoxide, which was inspired by the chemical structure of chlorpromazine.(20) The tricyclic antidepressants were developed through clinical trials, after chlorpromazine's chemical structure inspired their synthesis.(11) The monoamine oxidase inhibiting antidepressants were discovered in a clinical context, when they were noted to result in euphoria during use in tuberculosis treatment.(21)
Writing in the Comprehensive Textbook of Psychiatry, psychiatrist John Davis sums up the case of psychotropic drugs: "Many of the psychotropic drugs were discovered by chance when they were administered for one indication and observed to be helpful for an entirely different condition. The history of the development of both the antidepressants and the antipsychotic drugs points up the fact that major medical discoveries can evolve as a consequence of clinical investigation, rather than deductions from basic animal research."(13)
In the case of newer psychotropic medications as well, animal models have played neither a generative nor a crucial developmental role. In the case of bupropion, "a clinical hypothesis prompted the animal modelers' work."(11) With fluoxetine, in vitro studies demonstrated its selective inhibition of serotonin uptake.(22) While the standard animal tests failed to predict anti-depressant effects, human clinical trials demonstrated its value for depression therapy.(23)
The development of lithium as a psychotropic illustrates such clinical discovery. Although some proponents of animal experimentation have promoted the view that John Cade's discovery of lithium's anti-manic effects was preceded by crucial animal experimentation,(17,24) Cade's work with animals actually followed this discovery. Cade first learned of lithium's anti-manic effects in 19th century clinical reports of patients whose "gouty mania" was relieved by lithium salts.(24-26) The subsequent animal experiments merely dramatized the clinical findings and "sold" them to skeptical scientists.(11,15,27)
Data from animal model work has, in fact, been notoriously misleading in our efforts to understand psychoactive drugs. Animal experiments failed to identify the addictive potential of pentazocine (the use of which was eventually curtailed), the amnesic effect of triazolam (which underwent an altered dosage recommendation after years of clinical use), the abuse/addictive potential of methaqualone (eventually withdrawn), the overdose dangers of glutethemid (eventually withdrawn), or the lethality of nomifensine (eventually withdrawn).
In the case of anxiolytics, the abuse potentiat of the benzodiazepines was not revealed in animal studies. It took years of clinical use to reveal the substantial addictive properties of this class of drugs, which were due to their tendency to induce tolerance and dependence in chronic users. For example, Alprazolam is particulary addictive, in part because its duration of action is short. Animal studies, however, did not predict alprazolam's highly addictive potential.
Animal models have contributed little or nothing to the development of psychotropic medications. There is no reason to think that Barren's use of the animal model in an effort to discover a new type of anxiolytic will prove an exception.
Considering all the drawbacks and inadequacies of the animal model, why does such an experimental method continue to be proposed and funded? Barren himself provides a partial explanation with his comment that this method has been "extensively used . . . for studying anxiolytic drugs."(1) It is common for proponents of animal models to justify their use on the grounds that they have been used in the past. Given today's sophistication in computer manipulation and mathematical modelling,(28) and the degree of success of in vitro and clinical research methods in developing psychotropic drugs, Barren's nonhuman animal behavioral paradigm seems both archaic and inexcusable.Scientific Problems with the Animal Model
The animal model approach to drug investigation is inherently invalid; experiments on nonhuman animals do not yield data that can be extrapolated to humans.(29-31) Animal models differ from human conditions in several critical respects, including their means of induction, their natural histories, and, most fundamentally, in the species in which the condition occurs, Consequently, there are differences between the animals' condition and any "analogous" human condition in anatomy, physiology, and pathology that make animal data difficult, if not impossible, to interpret in terms of human disease.
Problems with animal models are perhaps most obvious in psychological research. Difficulties communicating with nonhuman animals undermine all efforts to understand animal psychology. Laboratory studies are particularly inappropriate, because it is impossible to interpret the psychological status of animals whose behavior is altered by the unnatural laboratory environment and whose normal psychology is profoundly altered by experimental manipulations that have no meaningful correlation to the natural environment in which the animals evolved. Furthermore, Barren's lever-pressing, tail-shocking tests illustrate these problems well. It is impossible to guage NO-328's precise effect on the monkeys. If the monkeys given the drug show increased lever-pressing, is this because the drug has acted as a pain-killer? An appetite stimulant? A tranquilizer? Has the drug merely induced hyperactivity? These questions can't be answered.
Further, animals experience hormonal, immunological, and neurological changes caused by the stress of being in a laboratory; these changes alter measurements and confound results(32)--precluding the discovery of useful findings. In monkeys and other primates, immobilization stress alone "has repeatedly been shown to have major endocrinological and physiological repercussions for the organism, including autoimmune disorders."(33) Barren's experiment creates severe distress by isolating inherently social animals, restraining them in a chair, shocking them, and traumatizing them with the insertion of a nasogastric tube. The pain and distress level induced by these conditions, which Barrett himself rates as "3" on a scale of one to five,(1) would produce sufficient changes in enzymatic, antibody, and neurotransmitter functions to render any obtained data invalid.
Conclusion
Barren's lever-pressing animal "model" cannot assess the clinical efficacy of anxiolytic drugs. His research illustrates difficulties with animal models in general and animal use in psychology in particular. Progress in the pharmacological management of anxiety will occur where advances in psychology have always occurred -- in the clinical setting, where careful observation of drug effects on human mood and mentation have been responsible for psychiatric drug discoveries.References
1. Barrett JE (principal investigator). Experimental Assessment of No-328: A GABA-Uptake Inhibitor With Potential Anxiolytic Activity. Federal grant application 018866-01.
2. Salzman C. The APA Task Force report on benzodiazepine dependence, toxicity and abuse (editorial). Am J Psychiat 1991;148:151-l52.
3. Sepinwall J, Cook L. Behavioral pharmacology of antianxiety drugs, in Iversen LL, Iversen SD, Snyder SH (eds). Handbook of Psychopharmacology, Vol. 13. Plenum Press, New York, 1978, pp 345-393.
4. Barrett JE. Behavioral pharmacology of the squirrel monkey, in Rosenblum LA and Coe CL (eds): Handbook of Squirrel Monkey Research. Plenum Press, New York, 1985, pp 315-348.
5. Keehn JD. Animal Models for Psychiatry. Routledge & Kegan Paul, London (Methuen, New York), 1986.
6. Sitaram N, Gershon S. From animal models to clinical testing -- promises and pitfalls. Prog Neuro-Psychopharmacol & Biol Psychiat 1983;7:227-228.
7. Cohen, MJ. The irrelevance of animal experimentation in modern psychiatry and psychology, in Natelson NB, Cohen MJ (eds): Proceedings of The First International Medical Conference "Future Medical Research Without The Use of Animals: Facing The Challenge." (Held in Tel Aviv, Israel, May 15-16, 1990.) Concern for Helping Animals in Israel, Alexandria, Virginia, 1991.
8. Cohen, MJ. Animal testing (letter). Psychiatric News, November 20, 1987.
9. Barnes D. The use of nonhuman animals in psychobiological and behavioral research, in Natelson NB, Cohen MJ (eds): Proceedings of The First International Medical Conference "Future Medical Research Without The Use of Animals: Facing The Challenge." (Held in Tel Aviv, Israel, May 15-16, 1990.) Concern for Helping Animals in Israel, Alexandria, Virginia, 1991.
10. Barnes D. Animal models (letter). Nature 1987;329:666.
11. Reines, BP. Psychopharmacologic discovery: The relative contributions of clinical and laboratory studies. Perspec Med Res 1990;2:l3-26.
12. Reines BP. Animal rights and research (letter). Am J Psychiat l988;145:539-540.
13. Davis JM. Antipsychotic drugs, in Kaplan HI, Sadock BJ (eds): Comprehensive Textbook of Psychiatry, Fourth Edition, Williams & Wilkins, Baltimore, 1985.
14. Campion FD. The AMA and U.S. Health Policy since 1940. Chicago Rev Pr, Chicago, 1984.
15. Reines B. The process of medical discovery, in Natelson NB, Cohen MJ (eds): Proceedings of The First International Medical Conftrence "Future Medical Research Without The Use of Animals: Facing The Challenge." (Held in Tel Aviv, Israel, May 15-16, 1990.) Concern for Helping Animals in Israel, Alexandria, Virginia, 1991.
16. The Use of Animals in Biomedical Research and Testing. Foundation for Biomedical Research, June 1988.
17. Pardes H. "Dear Colleague" letter sent to all American Psychiatric Association members, November 1989.
18. Baldessarini RJ. Antipsychotic agents. The Psychiatric Times, Jan 1988.
19. Krasney JK. Some thoughts on the value of life. Buffalo Physician 1984;18:6-13.
20. Anon. Serendipity in discovery of benzodiazepines. The Psychiatric Times, Sep 1986.
21. Lehman H, Kline NS. Clinical discoveries with antidepressant drugs, in Parnham MJ, Bruinvels J (eds): Discoveries in Pharmacology: Volume I, Psycho- and Neuropharmacology. Elsevier, New York, 1983.
22. Lemberger L, Farid NA, Bergstrom RF, Wolen RI. Fluoxetine, pharmacology and physiologic disposition. Inter J Obesity 1987;11(Suppl 3):157-161.
23. Lemberger L, Terman 5, Rowe H, Billings R. The effects of nisoxetine (Lilly compound 94939), a potential antidepressant, on biogenic amine uptake in man. Br J Clin Pharmacol 1976;3:2l5-220.
24. Cade J. Lithium salts for the treatment of psychotic excitement. Med J Austr 1949;36:349-351.
25. Johnson FN. The History of Lithium Therapy. London, Macmillan, 1984.
26. Reines BP. The relationship between laboratory and clinical studies in psychopharmacologic discovery. Perspec Med Res 1990;2:13-26.
27. Reines BP. On the locus of medical discovery. The Journal of Medicine and Philosophy 1991; 16:183-209.
28. Bross ID. Mathematical models vs. animal models. Perspec An Res 1989; 1:83-108.
29. Reines BP. A critique of animal psychology research at the University of California at Berkeley. Perspec An Res l989;1:25-55.
30. Bannister D. The myth of physiological psychology. Bull Br Psychol Soc 1968;21:229-231.
31. Shuttlesworth DS, Neil D, Ellen P. The place of physiological psychology in neuroscience. Physiological Psychology 1984;12:3.
32. Barnard N, Hou S. Inherent stress -- the tough life in lab routine. Lab Animal 1988;17:21-27.
33. Katsiia GV, Todua TN, Gorlushkin VM, et al. Effect of immobilization stress on gonadotropic function of the hypophysis in male hamadryas baboons (Papio hamadrysas). Biull Eskp Biol Med 1989;107:231-234.
[Note: Barrett continues his work today:
Khawaja X, Xu J, Liang JJ, Barrett JE. Proteomic
analysis of protein changes developing in rat hippocampus after chronic antidepressant
treatment: Implications for depressive disorders and future therapies.
J Neurosci Res. 2004 Feb 15;75(4):451-60.