Schizophrenia as a Model of Dopamine Disfunction

(Excerpted from Drugs, Brains and Behavior - by Timmons and Hamilton)


Return to main Table of Contents of The Text Book



One of the defining characteristics of advanced organisms is the ability to make flexible, yet adaptive responses to environmental stimuli. These stimuli may arise from within the organism or impinge upon it from the outside. The resulting myriad of stimuli ranges in salience from the barely noticeable to the intense. The stimuli in the intense range are usually considered to be biologically significant, whether they originate within the organism or are encountered in the outside environment.

At any given moment, the organism is likely to be faced with many stimuli that could be acted upon, but in reality only a few become the targets of behavior. Psychologists have conceptualized this process as a system of drives and rewards. The particular combination of stimuli that arises from the outside world and from the physiology of the organism triggers brain activity that has two major effects: It energizes behavior and directs behavior. For example, if an individual has gone for several hours without food, the stimuli arising from inside the body produce an effect which can be labeled hunger. These stimuli may be intensified by external cues such as the position of the hands on a clock, a television advertisement for junk food, or other food related items. There will be an increase in activity, and, given the appropriate circumstances, this activity will be directed toward food items. The consumption of food is said to be rewarding, and reduces the stimuli that initially energized the search for food.

The schema outlined above reflects the operation of the reward system, and as discussed in previous chapters, this system is believed to rely upon activity of catecholamines in fibers that arise from cells in the midbrain and project to various forebrain regions. Normally, this system organizes behavior in a systematic fashion that not only enhances the organism's ability to survive, but also makes it easier for psychologists to formulate laws of behavior. When this system fails to operate normally, the behavior of the affected individual strays outside of the normal expectancies, and some sort of label is attached to indicate a problem in behavioral adjustment. The behavior of the individual may be energized at inappropriate times (or at an inappropriate level), or it may be directed toward goal objects that are inappropriate or even nonexistent. Let us consider a couple of examples before going into the main body of this chapter.

One of the behavioral disorders that has been linked to the reward system both in its symptoms and its treatment is hyperkinesis or hyperactivity. This disorder typically appears in early childhood and is considerably more likely to afflict boys than girls. These children rarely require institutionalized care, but present tremendous challenges to teachers and parents. The disorder is characterized by impressive and unceasing physical activity. The child typically begins the day early with loud interactions with parents and siblings, running through the house, bumping into things and breaking them, getting into fights, spilling food, and so on. In school, the child distracts others by refusing to stay seated, fails to complete projects, and generally performs poorly. Bedtime is no exception to the exaggerated activity, and several tuckings in are likely to be required before both child and caretaker collapse into sleep.

When placed in the context of the reward system outlined above, hyperkinesis certainly can be viewed as an increase in the energizing aspect of drives, but it might also be characterized as a lack of directed behavior. Indeed, the disorder is now officially referred to as attention deficit disorder. It is in this regard that one of the most paradoxical and effective treatments comes into the scene. The behavior of these children would seem to require no further stimulation, yet one of the most effective treatments is the administration of amphetamines. The amphetamines are known to facilitate the functioning of neurons that release catecholamines, and have been shown to enhance rewards or to serve as rewards themselves. The most popular interpretation of the effectiveness of amphetamines in treating hyperkinesis is that the treatment enhances reward and provides more direction to the behavior. In both hyperkinetic and normal children (and adults for that matter), amphetamines can increase the attention span, which tends to normalize the behavior of the hyperkinetic children (e.g., Zahn, et al, 1980).

Autism is another disorder that may involve a dysfunction of the reward system, although the links have been less direct than in the case of hyperkinesis. The disorder is characterized by withdrawal from other individuals and a failure to respond to many external stimuli. Stereotyped behaviors are common, and may include sitting and rocking, manipulating an object over and over, or insisting that specific routines (e.g., going to bed) be followed in ritualized detail. This is a very serious and complex disorder that includes language difficulties and severe learning disabilities, but at least some of the symptoms can be related to the reward system. These aspects of the disorder are virtually the mirror image of hyperkinesis. The energy of the behavior is below normal and narrowly directed to only a few stimuli, while ignoring many other stimuli that are relevant to the normal individual. Although there are no particular drug therapies that are useful, there is a link to the reward system through some of the animal experimentation that was discussed in previous chapters. In particular, some of the work of Harlow and associates (e.g., Harlow and Suomi, 1971) showed that isolation of infant monkeys from both their mother and their peers produced stereotyped motions, withdrawal from other individuals, neglect of external stimuli, and (placing it in the context of our present discussion) large decreases in brain norepinephrine levels.

One of the current views of this disorder, that it may involve an excess of endorphins, was put forth by Kalat (1978) solely on the basis of behavioral symptoms. Consistent with this model is the observation that treatment with naloxone, a blocker of the endorphins, can ameliorate the symptoms in some cases. The close interaction between the endorphin systems and dopamine systems may, ultimately, form the foundation for a better understanding of autism.


The schizophrenic disorder has played a pivotal role in the development of a system of classification of behavioral disorders. It has long been the prototype of the group of serious disorders of thought processes that are termed psychoses. Less serious disorders of life adjustment are termed neuroses. These are old terms, and their derivations seem curiously reversed in the light of current knowledge. The term psychosis means disease of the mind, which seems reasonable enough, except that the term neurosis means disease of the neuron, implying a more physical basis for the less serious behavioral disorders.

The terms neurosis and psychosis reflect, in their original meaning, a related dichotomy of classification, namely, organic vs. functional. These opposing views of the roots of mental disorders go far back into history (as cited by Stein & Wise, 1971): St. Augustine (ca. 400 AD) took the functional viewpoint and asserted that "There are no disorders that do not arise from witchery, and hence from the mind." More recently, Thudicum (1884) made the interesting proposal that "...many forms of insanity are caused chemically by poisons fermenting within the body." These are not idle philosophical notions, because in a very real sense, they set the course for effective treatment. In the pure sense, one would expect behavioral methods to work for functional disorders, while chemotherapeutic methods might be better for organic disorders. These assumptions are clearly overstated, because of the known interactions between behavior, environment, and brain, but they nonetheless provide broad suggestions for the types of treatment that are most likely to succeed.

Historically, there have been some major successes in the treatment of severe disorders based on an understanding of the organic causes. The mental deterioration that accompanies the advanced stages of syphilis has been substantially controlled by penicillin. Pellagra, a disease which starts as a skin disorder and eventually results in severe mental deterioration, can be controlled by correcting the niacin deficiency of the diet.

Perhaps the most dramatic success has been in the treatment (though not the cure) of schizophrenia. This disease afflicts more than one percent of the population and is both chronic and progressive. It accounts for more than half of the resident mental hospital population. As indicated in chapter 4, there was a steady climb in the number of permanently institutionalized patients in mental hospitals during the early decades of this century. This dramatic rise was primarily a reflection of the increase in population, perhaps inflated a bit because of increasingly favorable attitudes toward institutionalized treatment. Schizophrenia was a major contributor to this patient population, and there seemed to be no end in sight. Then, after the mid-1950's, the number of chronically hospitalized patients dropped precipitously. This reduction can be attributed to the development of new drugs, the most important of which was chlorpromazine. (Unfortunately, many of these patients do not receive adequate care outside of institutions and fall into settings, such as living on the street, where they no longer take the drug and relapse into the full symptoms of the disease.)

As indicated in chapter 4, chlorpromazine was developed by the French physician, Laborit, as an autonomic stabilizer. The impact of this drug as an antipsychotic agent went far beyond anybody's expectations. The drug allowed the patients to return to the care of their families, or to less continuous care facilities. Even disregarding the tremendous reduction in human suffering, the role of chlorpromazine can hardly be overemphasized. In 1965, chronic hospitalization cost an estimated $7,000 per year. Today's costs may be more than ten times that value. The yearly financial savings can be conservatively estimated in the hundreds of millions of dollars!

The effectiveness of chlorpromazine in the treatment of schizophrenia set the stage for the development of realistic models of organic causes of mental disease--causes that are based on known functions of the neuronal physiology. This chapter will trace some of the highlights of the search for a plausible, if not valid neuronal model for schizophrenia. The groundwork for this search lies not only in the analysis of drugs such as chlorpromazine, but also in a more careful consideration of the major symptoms of schizophrenia.

The term schizophrenia has undergone many misinterpretations since it was first introduced (cf., Bleuler, 1950). The disorder is so common and the symptoms are so bizarre that the terminology has become a part of the layman's vocabulary. The most unfortunate misinterpretation is the translation of the term schizophrenia into split personality. Although the root of the word does refer to a splitting, Bleuler saw the disorder as a fragmentation of normal processes: "Thoughts, feelings and actions are not related and directed by any unifying concept or goal...a fragmentation of the central integrative process." It is precisely this lack of directed or unified behavior that Bleuler used to define the primary symptoms of schizophrenia:

  • Disturbance of thought patterns

  • Disturbance of affective reactions

  • Autism or withdrawal

These primary symptoms, individually and collectively, represent a loss of contact with reality. The patients show very atypical responses to their social situations. They commonly make up words and sentence structures, their responses to verbal communication may bear little or no relationship to the topic at hand, and their behavior is frequently at odds with the environment around them. Their affective reactions are inappropriate to the situation, with laughing or crying at the "wrong" times, etc. In later stages of the disorder, the affective disturbance is characterized by a profound inability to experience pleasure. In all of these situations, the patients are likely to show a withdrawal from the people around them, and in a sense, produce and respond to their own stimuli rather than to the outside world. This withdrawal also accounts for the rather curious fact that most people do not know a person suffering from schizophrenia. Yet, the incidence of schizophrenia is almost as great as the incidence of twins, and most people know several sets of twins.

The primary symptoms outlined above are frequently accompanied by one or more secondary symptoms. The symptoms were termed secondary because they are not necessarily present in all cases of schizophrenia. Ironically, these symptoms are the ones that are commonly portrayed in literature as prototypes of mental illness, so there is a natural tendency to assign greater importance to these symptoms in defining schizophrenia. These secondary symptoms include:

  • Hallucinations
  • Delusions
  • Paranoia

The secondary symptoms also reflect a loss of contact with reality. The hallucinations are usually auditory, and may include voices that provide instructions, criticism, or praise-- but with little or no basis in fact. Delusions of identity may occur, with patients adopting the identity of a famous individual, or delusions of unique powers. On the other side of the coin, they may have delusions of persecution, or feelings that they are being controlled by unique powers. Given the bizarre nature of their behavior, some of the paranoia that these patients may experience will be real. People, in fact, will talk about them and be suspicious of them. Yet, the paranoia of some of these patients is as extreme as it is bizarre, and clearly is pathological in nature.

The florid symptoms of schizophrenia lead naturally to the question, What happened to this person to cause these symptoms to develop? In many cases, it may be possible to find environmental factors that contributed to the disorder, but the evidence is very strong that the disease of schizophrenia has a firm biological (organic) basis which, at the very least, predisposes the individual to develop the symptoms.


Distribution of Occurrence

One of the best indicators for a biological basis of schizophrenia is the independence from culture. The disorder is represented in all cultures (be they primitive or industrial, permissive or strict, 1920's or 1980's) at a rate of about one or two percent. (This rate may sound low, until you calculate the real numbers for your home town or consider that there should be something on the order of 400 cases in Bangor, Maine.)

There is also a tendency for the disorder to appear in early adulthood, to preferentially afflict those with an ectomorphic body type, and (curiously) to be much less likely in those suffering from rheumatoid arthritis or Parkinson's disease. None of these factors would seem to be important in terms of environmental events that might precipitate the disease.

Genetic Patterns

A second line of evidence for a biological basis of schizophrenia is the strong genetic pattern of occurrence (cf., Heston, 1970; Nicol and Gottesman, 1983). The strongest relationship occurs between monozygotic (identical) twins. In cases in which one member of a monozygotic twin pair has schizophrenia, there is a 46% chance that the other member also suffers from schizophrenia. Thus, in about one half of the cases, one of the twins does not develop schizophrenia, even though the genetic makeupis identical; but pathology is still the norm, since only about 13% are judged to be normal. Dizygotic or fraternal twins, like siblings, share about one half of the genetic information. When a sibling or one member of a fraternal twin pair has been diagnosed as having schizophrenia, the other member bears about a 10-15% chance of being afflicted with the disease. Similarly, the child of a schizophrenic parent has a likelihood of becoming schizophrenic that is about 15 times that of the general population. Although the relationship is not perfect, there clearly is a high correlation between the closeness of genetic relationship to a schizophrenic patient and the probability of developing the disorder (see Fig. 7.1).

A word of caution is in order concerning the interpretation of the experiments that purport to show the relationship between genetics and the incidence of schizophrenia (or any other characteristic). There is also a strong correlation between genetic relationship and environment. If the home environment causes one child to become schizophrenic, would it not also be likely to cause siblings (or especially a twin) to become schizophrenic? Should one be surprised that a parent suffering from schizophrenia raises a child with similar characteristics? Although these are valid criticisms, there are a few experiments that provide counter-arguments. In some cases, it has been possible to find diagnosed schizophrenics who have twins or siblings that have been raised since birth or early childhood in foster homes. The results of these studies show relationships that are comparable to those described above. The likelihood of developing schizophrenia is more predictable based on the mental health of the biological mother than on that of the foster mother.

One of the most spectacular studies of the genetic and environmental influences on schizophrenia is that of the Genain quadruplets (Rosenthal, 1963). These quadruplet girls were the subject of intense study over a period of many years. Although it was impossible to eliminate all uncertainty, it seemed likely that they were monozygotic. One of these children developed schizophrenic symptoms at an early age. All three of the remaining quadruplets developed symptoms ranging from frank schizophrenia to schizoid tendencies during later years. The age of onset and the severity of the symptoms were related to birth weight, the lightest birth weight showing the earliest and most severe symptoms. The father of the girls and his mother both suffered from schizophrenia. The mother of the girls had some problems (not the least of which were a schizophrenic husband and quadruplet daughters!), but was not diagnosed as schizophrenic. This case study is probably a perfect example of the conclusions that should be drawn about the role of genetics and environment in schizophrenia: It seems obvious that one's genetic heritage can provide a strong predisposition toward schizophrenia, but whether or not the disease is manifested may be determined by other factors such as the home environment, the prenatal environment, dietary factors, and even other genetically determined characteristics.

Drug Effects and Schizophrenia

When Laborit developed chlorpromazine, his major motivation was to find a drug that would stabilize autonomic activity. In particular, one that would stabilize the stress response centrally, blocking the autonomic nervous system's reflection of the brain's interpretation of the response. The drug had such profound effects in the reduction of the stress response that it was rather quickly adopted for the treatment of patients who were suffering from anxieties arising from real or imagined stressors in their lives. In the language of the times, it was a major tranquilizer.

Chlorpromazine, like any other drug, has multiple effects, and some of these effects are troublesome during the course of therapy. The presence of these so-called "side effects" almost always leads to a search for related compounds that might share the parent drug's therapeutic effectiveness while diminishing the undesirable side effects. Over the years a fairly large number of related phenothiazine compounds were introduced into the clinic. Although none of these compounds was demonstrably better than chlorpromazine in the treatment of schizophrenia, the large scale clinical use of the phenothiazines set the stage for determining the most likely mode of action of these compounds-- namely, the blocking of dopamine receptors. The remarkably close correspondence between clinical potency and affinity for the dopamine receptors has already been seen in Figure 4.11. Correlations can be misleading, but the orderliness of these data lead, almost inescapably, to the conclusion that the phenothiazines produce their beneficial effects in schizophrenia by blocking the dopamine receptor.

The idea that schizophrenia is related to a dysfunction of the brain's dopamine system is consistent with several types of observations of drug actions. One line of evidence concerns the specificity of the phenothiazines in the treatment of schizophrenia. As we saw in Chapter 4, both the phenothiazines (e.g., chlorpromazine) and the benzodiazepines (e.g., Librium) have sedative or anxiety reducing properties. Despite this similarity in the treatment of these symptoms, Librium and related compounds are not effective in the treatment of schizophrenia. The benzodiazepines have little or no effect on dopamine receptors. Furthermore, chlorpromazine and related compounds seem to act rather specifically on the primary symptoms of schizophrenia, rather than the secondary symptoms (see above). The drugs that are most effective in treating these primary symptoms also tend to be the most likely to produce side effects that are similar to Parkinson's disease, a movement disorder that has been attributed to a deficiency of dopamine in certain brain areas.

A second line of evidence that supports the link between primary symptoms of schizophrenia and the dopamine system are the effects of hallucinogenic drugs. Compounds such as LSD produce hallucinations in both schizophrenic and normal individuals. However, these drug effects are usually distinct from the schizophrenic symptoms, and the schizophrenic patients recognize the hallucinations, but do not attribute them to their illness (cf., Hollister, 1962). These observations suggest that the hallucinations, which Bleuler considered to be a secondary symptom of schizophrenia, may involve non-dopamine systems that are not involved in the primary disorder.

The effects of amphetamine, a drug which stimulates neurons that release dopamine (as well as those that release norepinephrine), also strengthen the case for dopamine involvement in schizophrenia. In normal individuals, small dosages of amphetamine produce psychomotor stimulation. Repeated high dosages of amphetamine produce greater stimulation, and in some cases, a breakdown that is indistinguishable from acute paranoid schizophrenia. As it turns out, the diagnostic distinction between an amphetamine overdose as opposed to a schizophrenic breakdown is not immediately critical. In either case, the treatment of choice is the administration of chlorpromazine or some related phenothiazine. This line of argument can be extended by observing the effects of amphetamine on schizophrenics. Low doses of amphetamine that would produce only a mild psychomotor stimulation of a normal individual, produce a strong and specific activation of the primary symptoms of schizophrenia (cf., Janowski et al, 1972). Unlike the effects of the hallucinogenic drugs noted above, the patient attributes the effects of the amphetamine to a worsening of the schizophrenic condition. Again, phenothiazines are effective in reducing these symptoms.


Searching for a Chemical Label

There have been hundreds of futile and misguided attempts to determine the biochemical basis of schizophrenia. Many of these experiments took the statistically unlikely approach of trying to find a unique chemical in the blood, urine, saliva, or some other body fluid of the schizophrenic patient. There were many positive results, but nearly all of these turned out to be false leads, with the apparently unique chemical substance being unrelated to schizophrenia, and instead, being the result of differences in physical activity, institutional diet, by-products of medications, or the amount of coffee consumed. This is clearly a needle in the haystack approach. Given the thousands of body chemicals, the likelihood of choosing to assay the one (if there is one) that is a unique marker for schizophrenia would be very small indeed.

A much more logical and positive approach to trace the causes of schizophrenia is to base the search on behavioral and neurochemical systems that are known to be involved in the disease of schizophrenia. A model of schizophrenia that is based on the function of a particular set of neurons would be much more plausible than one simply based on the presence or absence of some chemical.

The DBH Model


One of the most appealing neuronal models of schizophrenia is the so called "DBH Model" (dopamine beta hydroxylase) that was put forth a number of years ago by Stein and Wise (1971). The model is almost certainly wrong in at least some of its details, but it is exemplary because of its testability, its basis on known physiological processes, and the relationship of these processes to behavior. We will trace this model in some detail because portions of it may be correct, and because it serves as an excellent model of the type of approach that must be taken to find the roots of any organically based disorder of behavior.

The initial focus of the Stein and Wise theory was on the behavioral aspects of schizophrenia. The lack of organized and directed behavior, the inability to experience pleasure, and the withdrawal from the external environment all pointed to a dysfunction of the reward system. But what could be causing this deterioration of these neurons? Could it be that Thudicum had been correct many years ago when he suggested that insanity might be caused by some poison, fermented within the body? If so, what might the characteristics of that poison be? Given the clinical progression of schizophrenia, several limitations would have to be placed upon such a poison. It would have to be produced more or less continuously over the course of many years to account for the progressive deterioration of the behavior of schizophrenics. It would have to act rather specifically on certain populations of neurons to account for the primary symptoms of schizophrenia. It would have to be more or less immune to tolerance, or the body would quickly adapt to it, causing a short term disorder. Finally, it is likely that the substance would be present in relatively small quantities, or it would have been captured in the net of one of the many biochemical studies that have been done. What sort of a chemical might this be?

Stein and Wise proposed that the disease of schizophrenia might be caused by low levels of 6-hydroxy dopamine (6-OHDA) circulating in the bloodstream. As indicated in previous discussions (cf., Chapter 5), 6-OHDA is a powerful and specific neurotoxin that enters the terminals of catecholamine bearing neurons and physically destroys them. As shown in Figure 7.2, the administration of 6-OHDA to rats causes them to stop responding to the rewarding effects of brain stimulation, mirroring the inability of schizophrenics to respond to pleasure. Furthermore, these effects can be blocked by chlorpromazine and related phenothiazines. The most likely mechanism for this protective effect is the blockade of the specific receptors that are involved in the reuptake mechanism of these neurons. The inability of the 6-OHDA to enter the cells exposes it to other metabolic process that inactivate it, while protecting the cell from the destruction that would ensue if the 6-OHDA were allowed to enter. A final feature that mimics the disease of schizophrenia, is that the rats that have been poisoned with 6-OHDA show waxy flexibility, a bizarre dysfunction of the motor system that allows them to be "molded" into postures that are sustained for long periods of time.

But are these symptoms more than coincidental? Is it reasonable to suspect that the disease of schizophrenia is caused by a neurotoxin that was invented in a biochemist's test tube? If so, what would cause one or two percent of the population to produce the substance while the remaining population does not?

Despite the superficial similarities between the effects of 6-OHDA and the symptoms of schizophrenia, the model would seem to be implausible without some mechanism for the body to produce the neurotoxin. The suggestion of a mechanism by Stein and Wise was both ingenious and insightful. They proposed that the root of the disorder was an enzyme deficiency.

DBH as a rate limiter

Figure 7.3 shows the normal sequence of reactions in the formation of the catecholamine neurotransmitters. Each step of the sequence is assisted by a specific enzyme which, as a rule, speeds up the reaction. One of the characteristics of a sequence of biochemical reactions such as this is that the overall rate of transformation of the initial compound into the final product (in this case, the conversion of the amino acid tyrosine into norepinephrine) can proceed only as fast as the slowest step. By analogy, if water is being poured into a bucket through a series of funnels, the bucket will fill only as fast as allowed by the smallest funnel. In some cases, a sequence of biochemical reactions might be limited by the amount of compound available for conversion, or by the amount of enzyme present at one of the steps, or by the speed of the enzyme. In this particular example, it has been established that the so-called rate limiting enzyme is at Step 1, the conversion of tyrosine into DOPA (dihydroxyphenylalanine). The enzyme is called tyrosine hydroxylase, which describes the nature of the chemical action that it facilitates.

The location of the rate limiting reaction at Step 1 means that in the normal sequence of events, this step governs the remainder of the sequence. As in the case of an assembly line, the faster activity at Step 2 converts all of the DOPA into DA and must "wait" until additional DOPA is formed. Likewise, Step 3 is limited by the amount of DA that is formed. These phenomena are confirmed by biochemical experiments which show that modest increases or decreases of the enzymes at Steps 2 and 3 have no effect on the sequence of reactions. By contrast, small changes in the activity of tyrosine hydroxylase are reflected in the overall conversion of tyrosine into norepinephrine (see Fig. 7.4).

Stein and Wise proposed that schizophrenia involves a deficiency of the enzyme in Step 3, dopamine beta hydroxylase or DBH. If this enzyme is present in smaller than normal quantities, it can become the rate limiting enzyme and set the stage for the sequence of events shown in Figure 7.5. According to this scheme, tyrosine would be converted to DOPA, all the available DOPA would be converted to dopamine, but the dopamine would be present in excess because of an inability to rapidly convert it into norepinephrine. It is this excess of dopamine that allows abnormal chemical conversions to take place. The availability of dopamine, exposes it to the action of other enzymes. The details of these enzymes are not important in the present context, but Stein and Wise presented biochemical pathways that could convert dopamine into 6-OHDA. Once the body has formed this neurotoxin, it could be taken up by the neurons that normally release norepinephrine or dopamine, and the cells would be destroyed in the same manner as if the drug had been administered by a hypodermic needle. Thus, the metabolic machinery is there for Thudicum's poison to be "...fermented within the body," and Stein and Wise have shown how the machinery might be put into action when normal pathways of neurotransmitter production have gone awry.

There is also another line of evidence that can be supported by the DBH hypothesis. As a result of some of the earlier biochemical sleuthing, it was determined that many patients suffering from schizophrenia have a characteristic odor in their sweat. Rats, dogs, and even humans can be trained to discriminate between odors derived from the sweat of schizophrenics and that derived from normal subjects (Smith et al, 1969). This substance has been isolated and identified. Curiously, the substance can be formed in a few steps by beginning with 6-OHDA and proceeding through a series of biochemical reactions that can take place in the human body.

The DBH model of schizophrenia has been strongly criticized. The naysayers point out that 6-OHDA per se has not been found in schizophrenics. Nor has it been demonstrated that the 6-OHDA is the substrate that forms the identifiable odor substance. When scrutinized in fine detail, some of the similarities of the effects of 6-OHDA on rewarded behavior in rats seem only remotely related to the symptoms of schizophrenia. The biochemical assays of the levels of DBH in the brains of schizophrenic patients (performed at autopsy; Wise and Stein, 1973) showed lowered levels, but the results were not as dramatic as one might like. The model has, however, served as a focal point of discussion and of the design of experiments over the years. It is also of historical importance, because even if it does not turn out to be an accurate model for the disease of schizophrenia, it is an admirable model for models. It is precisely this type of neurological model, based on variations of firmly established phenomena, that will eventually unlock the mysteries of schizophrenia. The next section will discuss some additional proposals for the nature of schizophrenia. Some of these oppose the DBH theory, others are consistent with it, but none is as global in its approach.

Other Dopamine Models

Metabolic pathway theories

Whether or not the primary neuronal deficit in schizophrenia is a deficiency of DBH, it seems likely that some disorder of dopamine metabolism is involved. The detailing of this change in metabolism has remained elusive because of several problems: Foremost is the problem of diagnosis, with "schizoid" symptoms presenting themselves in a variety of mental disorders that may not turn out to be true schizophrenia, as well as the likelihood that several different forms of schizophrenia may exist (we shall return to these problems later in the chapter). The progressive nature of schizophrenia also presents a problem because some of the biochemical markers for the disorder actually may be more salient during the early stages than after the disease has become more serious. Finally, there remains the simple fact that these patients require care, and it is difficult for researchers to find a patient population that has not been treated with phenothiazines or other drugs that are known to produce large changes in the very systems being investigated. Despite these and many other problems, the research findings have shown a continuous, if not entirely coherent, thread of evidence for impaired function of the dopamine systems.

Wise and Stein reported in 1973 that the brains of schizophrenic patients had somewhat lower levels of DBH than normal persons of the same age and sex who had died in car accidents. More recently, this biochemical difference has reemerged with the development of techniques that are more sensitive than relying on autopsy data. Some schizophrenic patients have lower than normal levels of DBH in the cerebrospinal fluid, while others do not. Those who show this deficiency tend to respond better to phenothiazine treatment and also appear to have brain atrophy, as demonstrated with computer tomography techniques (Sternberg et al, 1982; van Kammer et al, 1983). Not only are these data consistent with the original hypotheses of Stein and Wise, but it has also been shown that phenothiazine activity can be enhanced by simultaneously administering alpha methyl tyrosine, a potent inhibitor of the normal rate limiting enzyme, tyrosine hydroxylase. This treatment, in effect, would allow tyrosine hydroxylase to regain its status as rate limiter, even in the face of lowered DBH levels.

A dysfunction of dopamine cells would also be expected to result in different levels of catecholamine metabolites by changing the relative involvement of the various enzyme pathways (cf., Fig. 6.13). One product that has been linked repeatedly with schizophrenia is (HVA) homovanillic acid, although the results are not as consistent as one might hope. Pickar and associates (1984) found elevated levels of HVA circulating in the plasma of schizophrenia patients. Phenothiazine treatment produced a gradual reduction of HVA levels over a 3- to 6-week period, and the improvement of schizophrenic symptoms showed a comparable delay. These data are very impressive, but it should also be noted that other investigators have reported normal HVA levels (e.g, van Kammer et al, 1983) that may increase with phenothiazine treatment (e.g., Bacapoulos et al, 1979). One possible resolution of this apparent discrepancy is that one group measured circulating levels of HVA in the plasma, while the other measured HVA levels of brain tissue. These different compartments may, in a sense, represent the same scene from different angles.

Amphetamine and amphetamine-like compounds have also continued to play a role in theorizing about schizophrenia. Phenylethylamine (PEA) has been found to be twice as high in the urine of paranoid schizophrenics than in either non-paranoid patients or the normal population (Potkin et al, 1979). This is interesting not only because low dosages of amphetamine can exacerbate the symptoms of schizophrenia, but also in light of recent findings of neurochemical changes that accompany chronic amphetamine administration. Trulson and Jacobs (1979) administered large dosages of amphetamine to rats over a three-week period. In addition to the expected decline in norepinephrine and dopamine levels, they also found a reduction in serotonin (5-HT) and the related compound, 5-HIAA. Hallucinogenic drugs are also known to interfere with 5-HT and 5-HIAA, and they produce the same abnormal behaviors of limb licking and abortive grooming that were observed in the amphetamine treated animals.

There is little doubt that dopamine metabolism is abnormal in schizophrenia patients, but we may be dealing with a chicken and egg problem: Is the metabolic change the cause of the disease or does it reflect a reaction to some more primary cause such as a change in receptor number or sensitivity? We turn now to a discussion of some of the research findings on receptors which may help to confirm that the metabolic changes are, in fact, the chicken (or the egg).

Receptor theories

We already have pointed out the powerful interplay between the clinician and the neurochemist in demonstrating the relationship between phenothiazine potencies and their ability to block the dopamine receptor (cf., Fig. 4.11). This relationship has become even more impressive with the introduction of more drugs and the discovery of different dopamine receptor types (D1, D2, and maybe even D3 and D4). The relationship between clinical potency and the ability to block the D2 receptor is nearly perfect, while the affinity for the D1 receptor is virtually unrelated to clinical efficacy (see Fig. 7.6). The question then becomes more specific: What does the D2 receptor do? The current evidence is that this receptor is involved in neurotransmitter functions of dopamine that are related both to schizophrenia and to the extrapyramidal motor system. But the key to the importance of the D2 receptor may be its neuromodulatory function rather than direct neurotransmission.

Dopamine receptors appear to mediate both the transient neurotransmitter functions (e.g., depolarization) and the longer term neuromodulatory effects that alter cell metabolism (Figure 7.7). An understanding of these dual effects requires the introduction of a new concept, that of second messengers. The neurotransmitter is the first messenger, and its arrival at the receptor carries the information that the presynaptic cell has been active (in this regard, the receptor on the post-synaptic membrane is now coming to be referred to as the recognition site). The recognition site may simply confirm that the presynaptic cell has sent a signal. This information must then be translated into a cellular response. Typically, we think of this cellular response as being short-term in nature, reflected as a depolarization that may initiate an action potential to continue the message on down the line. In other cases, however, the result might be a long-term change in some aspect of the cell's physiology such as a change in the number of receptor sites or a change in the rate of transmitter production or storage. In all of these cases, it appears that second messengers are involved: the recognition site triggers some second event that may alter an ion channel (in the case of typical depolarization) or alter the structures of proteins that are involved in enzyme systems or receptor structures. The two most common second messenger systems are cyclic AMP and the phosphoinositide cycle. Both of these are involved in the changes in protein structure that mediate both short- and long-term changes in the neuron. In this particular example, the interaction of dopamine with the D2 membrane receptor inhibits the second messenger, adenyl cyclase. The adenyl cyclase mediates the conversion of the high energy compound ATP into cyclic AMP. This provides the energy for the formation of protein kinase, which is involved in the phosphorylation reactions for the synthesis of the various enzymes of the cell.

Although this whole process begins to sound a little like the house that Jack built, the end result is that the release of dopamine onto these receptors can control the metabolism of the cell, determining for example, the rate of synthesis of the neurotransmitter (which may or may not be dopamine in this postsynaptic cell). Returning now to the specific receptor types, the interaction of dopamine with the D2 receptor inhibits adenyl cyclase, while the D1 receptor enhances it. If we assume an excess of dopamine activity in schizophrenia, then the role of phenothiazines is not only to block the depolarizing effects of dopamine on the postsynaptic cell, but also to allow the metabolic functions of the postsynaptic cell to return toward normal by interfering with the abnormally high inhibition of the second messenger, adenyl cyclase.

Endorphin contributions

The endorphins (cf., Chapter 7) have also entered the scene in the attempts to provide a biochemical model of schizophrenia. The early reports of the alleviation of schizophrenia symptoms by administering (or blocking) endorphins were probably more enthusiastic than real. But as the understanding of these neuropeptides improve, they will almost certainly play a major role in the treatment of schizophrenia. In addition to serving as neurotransmitters, endorphins have at least two major influences on the dopamine systems: They can modulate the release of dopamine by acting on presynaptic receptors, and they can change the number or sensitivity of dopamine receptors on the postsynaptic membrane by influencing the protein phosphorylation mechanism described above (Volavka et al, 1979; see Fig. 7.8). Some of these effects may be mediated directly by neighboring peptidergic neurons. Others may involve the release of peptides along with dopamine.

But what about Dale's Law of one transmitter per cell? One of the last bastions of simplicity in brain models, this law has been rather thoroughly repealed. Alas. The neurochemical maps of the brain now include a column for coexisting peptides, and the presence of multiple transmitter substances seems to be more the rule than the exception (e.g., Hoekfelt et al, 1984, 1987). It may still be true that only one compound serves the specific neurotransmitter function of changing the polarization of the membrane, but the other ingredients of this synaptic cocktail have important effects on such things as the regulation of release, the sensitivity of receptors, and the long-term changes in metabolic functions of both the presynaptic and postsynaptic cells (Fig. 7.9 ).

The dynamic synapse

Another important aspect of neuromodulation is the regulation of receptors. Receptors and transmitters are not static entities in the lock and key sense, but rather they are in a state of dynamic equilibrium that has been likened to the concept of homeostasis in other physiological systems. The membranes involved in the release and reception of neurochemicals have a real half-life, which is another way of saying that they wear out and must be continually replaced. We have seen mechanisms for changing the synthesis, release, autoreception, postsynaptic reception, and second messenger effects in the next cell. Normally, these are all orchestrated to meet the current exigencies of a neural system. When this complicated set of controls goes awry, a disease state such as schizophrenia may result.

This dynamic model of the synapse also adds several levels of complexity to the actions of drugs, as we have already seen in earlier chapters. The time course of drug effects may provide a clue concerning the nature of the therapeutic effect. The phenothiazines have some immediate effects, such as the calming of agitation, but the relief of many of the more definitive symptoms of schizophrenia may not appear for several weeks. This delayed action (see also the discussion of delay of antidepressant drugs) suggests that the effects rely upon long term metabolic changes that are triggered by the drug. One of the more detailed models of the therapeutic effects of phenothiazines has been developed by Bunney and associates (e.g., Bunney, 1984; see Figure 7.10). According to this model, phenothiazines initially block inhibitory feedback, increasing DA release. This results in the formation of more DA receptors, but postsynaptic activity is eventually blocked (after a few weeks) by a combination of direct phenothiazine blockade and chronic depolarization. The complexity of this model goes well beyond the level of this text, but the principal is of central importance: The therapeutic effects of drugs cannot always be fully appreciated by their simple actions of blocking, mimicking, and other synaptic actions. These actions cause reactions, and the cell systems that are influenced by the drug make long-term changes to reflect the exposure to these effects.

Movement Disorders

Movement disorders have a long history of association with schizophrenia. One of the early observations was the mutually exclusive nature of Parkinsonism and schizophrenia. Each disorder afflicts a sizeable portion of the population and, by chance, one would expect 1-2% of the patients with Parkinson's disease to be afflicted with schizophrenia. Such patients are rare, and some investigators would claim that clear-cut cases are nonexistent. Why? As the understanding of both diseases progressed, dopamine became the focal point of both diseases, with Parkinsonism being characterized by dopamine deficiency and schizophrenia as dopamine excess. More recently, the emphasis has shifted to receptors, with Parkinsonism being linked to too few receptors (perhaps the result of autoimmunity; cf., Chapter 7) and schizophrenia to too many receptors.

Although both diseases are linked to dopaminergic systems, and even to the D2 receptors, the anatomical substrates may be separable (Fig. 7.11). There are two, rather distinct clumps of dopaminergic cell bodies (known to anatomists as A9 and A10) in the region of the pons. The A9 region is roughly equivalent to the substantia nigra, and projects fibers to the striatum via the nigrostriatal pathway, forming the core of the extrapyramidal motor system. In close juxtaposition to this system, fibers from the A10 region (ventral tegmental area) project to various portions of the limbic system and some cortical regions, most of which have been linked to emotional responsiveness and-- when in dysfunction-- to schizophrenia.

We have already seen that the phenothiazines block the D2 receptor, so it should come as no surprise that a major "side effect" of these drugs is an impairment of extrapyramidal functions. In fact, the symptoms produced by these antipsychotic drugs are similar to those that occur in the disease state of Parkinsonism. The short term problems with these side effects are not insurmountable. The dosage of phenothiazines can be adjusted to minimize these effects, and some degree of tolerance appears to develop to the extrapyramidal effects. Furthermore, the extrapyramidal effects can be alleviated by the administration of antimuscarinic drugs, or by choosing a phenothiazine that happens to also have muscarinic blocking properties. In the long run, however, the prognosis is somewhat bleak. If the neuromodulatory effects triggered by the phenothiazines can gradually reduce the symptoms of schizophrenia, there is no reason to suspect that they would not also gradually produce extrapyramidal difficulties. Indeed, it has been demonstrated that haloperidol produces neuronal sprouting within the nigrostriatal system (Benes et al, 1979), presumably in counter-response to the DA receptor blockade. Clinically, these effects are exhibited as a troublesome and largely irreversible decline in extrapyramidal function which has been labeled tardive dyskinesia (the term tardive referring to the delayed development of the disorder). Tardive dyskinesia is, in some sense, the exact opposite of Parkinson's disease. Most clinicians feel that the risk of progressive motor dysfunction is a substantial, though acceptable, price to pay for the control of the progressive and even more debilitating symptoms of schizophrenia. Unfortunately, there is at present no apparent neurochemical distinction between these two systems that allows specific treatment of schizophrenia: The phenothiazines remain the drugs of (reluctant) choice.

We alluded earlier to the clinical complexity of schizophrenia, a disorder that presents a baffling array of symptoms to the clinician. Furthermore, there are other types of disorders that present symptoms that are also characteristic of schizophrenia. Two of the more notable mimickers of the disorder are amphetamine overdose and the manic phase of bipolar affective disorders. The manic disorders have sometimes been termed schizoaffective, an unfortunate label, because it encourages the prescription of phenothiazines. Taylor (1984) has given an insightful discussion of this problem and provides guidelines for diagnosis and treatment. In diagnosis, family histories are important because the genetic link provides additional clues to the likelihood that the disorder may be an affective disorder rather than schizophrenia. Furthermore, the diagnosis of schizophrenia on the basis of agitated or bizarre behavior is unreliable, because these behaviors also typify manic disorders. Taylor admonishes clinicians to adhere to an old rule of thumb: When a diagnosis is in doubt and the patient may be suffering from one disease or another, always treat the disease that is easier to cure. In this instance, the affective disorder has the best prognosis. A course of treatment with lithium is very likely to alleviate the manic disorder. If the diagnosis was wrong, treatment with phenothiazines can begin. If the clinician chooses schizophrenia and begins the chronic course of treatment with phenothiazines, the patient is exposed to the high risk of extrapyramidal damage. Taylor points out that schizophrenia is the diagnosis for 4-8% of admissions, whereas the accepted incidence of schizophrenia is about 1-2%; he suggests that permanent brain damage is too high a price to pay for this high rate of misdiagnosis.



1. The reward system normally functions to energize and direct behavior toward appropriate environmental stimuli. Both norepinephrine and dopamine appear to be involved.

2. Schizophrenia may result from a disturbance of this system, producing the symptomatic changes in thought patterns, inappropriate affect, and social withdrawal.

3. The consistent rate of occurrence in all cultures and the apparent heritability of schizophrenia strongly suggest a biological basis for the disease.

4. Chlorpromazine and related phenothiazines are powerful antagonists of schizophrenic psychoses.

5. Drugs that are most potent in treating schizophrenia are also the most effective blockers of dopamine receptors. These results suggest that schizophrenia may be the result of excessive dopamine activity.

6. The DBH model suggests that an enzyme deficiency causes dopamine to accumulate and then be transformed into a neurotoxin, 6-hydroxy dopamine. This neurotoxin destroys noradrenergic and dopaminergic terminals.

7. Other theories do not agree with the neurotoxin notion, but nearly all suggest some type of abnormality within the dopamine systems.

8. There is strong evidence that schizophrenic patients have an excess number of dopamine receptors.

9. Chlorpromazine and other drugs that are used to treat schizophrenia typically require several weeks to reach full effectiveness. This suggests that they may act by triggering neuromodulatory processes.

10. Endorphins are likely to be involved as co-transmitters with dopamine and as neuromodulators to change the number or sensitivity of dopamine receptors.

11. The extrapyramidal motor system involves closely related dopamine fibers.

12. Parkinsonism is a movement disorder that involves a deficiency of dopamine receptors, and patients suffering from this disease rarely if ever develop schizophrenia.

13. Drug treatments of schizophrenia often produce Parkinson-like symptoms which may be, in some cases, irreversible.



Adenyl cyclase

Amphetamine psychosis

Attention deficit disorder



Cyclic AMP

D1 and D2 receptors


Dynamic synapse


Extrapyramidal system





Nigrostriatal pathway




Phosphoinositide cycle

Primary symptoms

Rate limiting enzyme

Reward system

Schizoaffective disorders


Second messenger

Tardive dyskinesia

Tyrosine hydroxylase




   Copyright 1996-2004. All Rights Reserved.