Chapter 5




Pain Pathways and Measurement of Pain

The Discovery of Opiate Receptors

Pain Reduction Systems

Behavioral Effects on Pain Reduction

Interpretation of pain.

Placebo and acupuncture.

An Overview of the Pain Response


Survey of the Immune System

The biological self.

Humoral responses.

Cellular responses.

Behavioral Effects on the Immune System

Interpretation of the environment.

Learned immune responses.

Interaction with endorphins.

Implications for receptor function.

Autoimmunity and behavioral disorders.





Return to main Table of Contents



There is nothing that rivets the attention of an organism quite like pain. It is the first line of defense against environmental situations that threaten the existence of the individual. Each of the sensory modalities has upper limits at which intense stimuli become painful. Although we can recognize the difference between a painfully bright light and a painfully loud sound, there is a commonality among intense stimuli that transcends the modality, placing loud noises side by side with ingrown toenails, aching teeth, and of course, electric shock.

There are many factors that influence the quality and quantity of pain, the most obvious being the physical intensity of the stimulus that is producing the pain. But other parameters also can have powerful influences on the interpretation of pain. The acute pain produced by hitting a thumb with a hammer may be less bothersome than the milder throbbing of a chronically injured knee. Mowrer (1956) put forth a model of fear which, among other things, postulated that the aversive quality of an intense stimulus included both pain and fear of more pain. Mowrer's theory was complicated and difficult to support, but it presaged by a couple of decades the important distinction between predictable and unpredictable pain.

The involvement of pain in a wide range of human conditions has led to all sorts of remedies for the reduction of pain. Some of these are behavioral and almost reflexive in nature, such as sucking on a thumb that was hit or grasping a barked shin with the hands. This application of pressure may have some physiological basis for pain reduction because of the stimulation of alternative pathways that may actively compete with the processing of pain information. Along the same lines, we may even bite a knuckle to reduce pain in the foot or, in the frontier tradition, bite a bullet instead of a knuckle. The application of cold or heat is also widely prescribed as a physical means of reducing pain.

The ancient folk remedies also include the pharmacologic reduction of pain, the most notable of which are alcohol and the opiate compounds. As modern pharmacology began the systematic search for drugs, the anesthetic compounds (both local and general) were developed. Interestingly, there was some reluctance to use these compounds, partly because of the unknown actions of these drugs, and partly because of the notion that pain was an important part of the healing response. These objections were soon cast aside, however, and drugs that offer pain relief now comprise a major portion of both the prescription and nonprescription pharmaceutical industry.

The most exciting developments in pain research during recent years have not been in the discovery of drugs, but rather in the emerging story of how the body reacts to pain. As indicated above, pain is essential to allow the organism to minimize exposure to adverse environments. But once this alerting function has been accomplished, there is a diminished need for a continuation of the painful stimulus. (It is not necessary to continue feeling the full impact of the hammer on one's thumb to make one more careful in the future!) Toward these ends, it seems that there are mechanisms for the reduction of continued pain. These findings will form the foundation for the present chapter.


Pain Pathways and Measurement of Pain

The distinction between acute pain such as that produced by a pin prick and the more chronic, sometimes throbbing pain that follows is recognized readily on an experiential level. These have been referred to as 1st pain and 2nd pain, respectively, and there is evidence that these differing perceptions may have an anatomical basis, with the 1st pain being mediated by larger faster fibers that travel through the lateral spinothalamic tract and 2nd pain being mediated by smaller, slower fibers that pass through the medial spinothalamic tract. Because of this anatomical distinction, it has been possible to reduce chronic pain surgically by the transection of the 2nd pain fibers in the spinal cord or, in some extreme cases, by lesions directed to the thalamic targets of these fibers. In the latter case, some patients have reported that the pain is still present, but they "do not care", suggesting a separation of the emotional aspects from the detection of pain.

The investigation of pain (and its counterpart, pain relief) requires the systematic and quantitative measurement of the phenomena. In humans, this has been approximated by measuring thresholds of reported pain under carefully controlled conditions. These thresholds are not discrete, but can be easily influenced by prior experience, instructions, social expectations, and a variety of other factors. Although this complicates research, it does not necessarily mean that the measure is faulty, but probably reflects the very real changes in the perceptions of pain that are engendered by these conditions.

A number of different procedures have been developed to assess the pain threshold of experimental animals. One of the more interesting of these is the flinch/jump test, which exposes a rat to several series of shocks, in ascending and descending intensities (Evans, 1961). With increasing intensities, for example, there will be a range of low shock levels to which the rat does not respond. Then, as the intensity increases, the rat will begin to show a slight flinch with each brief shock presentation. With further increases in intensity, the rat will actually jump when the shock is presented, the criterion usually being that at least three of the rat's feet leave the grid floor. Once these responses have been determined, the sequence is reversed and descending shock intensities are delivered until the jump response disappears and, with further decreases, the flinch response disappears. This procedure produces reliable threshold determinations for both the flinch response, which is interpreted as the lowest shock level that is detectable as pain, and the jump response, which is interpreted as the lowest level of shock that produces an emotional response to the pain. The test can discriminate, for example, between a drug that locally blocks nerve conduction, and a centrally active analgesic drug that reduces the impact of the pain (see Fig. 5-1). (Some of the readers may have had pain thresholds established by a dentist to determine the relative health of two or more teeth, and will be able to appreciate this difference between simple detection and an emotional response.) Despite the theoretical advantages of the flinch/jump test, this procedure has not been used routinely, because it requires an observer to make a subjective evaluation of whether a particular response is a big flinch or a small jump--not as easy as it might seem. Furthermore, the test is rather tedious and time consuming to administer.

Two simpler tests, both of which use heat as the pain eliciting stimulus, have been used more frequently. One of these is the paw lick test, which involves placing the rat on a specially constructed metal plate which is maintained at a constant temperature. The temperature is set low so that the rat can remain on the plate for several seconds before it becomes painful (something akin to the handle of a skillet, which may seem only warm at first, but nonetheless add a briskness to one's steps when carrying it across the kitchen.) The measure of the pain response is the latency from the time the rat is placed on the plate until it licks its front paw. An even simpler measure is the tail flick response, which can be automated for objective measurement. In this test, the rat (or mouse) is placed in a small restraining cage and its tail is pressed lightly into a groove. A source of heat (usually a light bulb) is directed to the underside of the tail. When the heat reaches the pain threshold, the tail is flicked out of the groove, and the latency between the onset of the heat and the tail flick is automatically recorded. Both of these tests produce reliable measurements of thresholds, and have become the standard tests for determining the analgesic properties of drugs or behavioral treatments (see Fig. 5-2).

We turn now to a discussion of the research that has focused upon analgesia, the relief of pain.

The Discovery of Opiate Receptors

Opium, an extract of the poppy plant, was ensured a place in history many centuries ago through the writings and the art work of early civilizations. Loosely translated, the term narcotic means "numbing" and probably refers both to the direct analgesic properties of this compound and to the more general depressant or sedative properties of the drug in larger dosages. Because of its unique powers and potential for abuse, opium and its derivatives have been the subject of literature, art work, legislation and even wars. Against this backdrop of human drama, a research story has unfolded, the results of which may have more far reaching consequences than all of these other aspects.

The production of opium as a drug was mastered long before there existed any formal knowledge of pharmacology. The harvesting of the poppy pods and the procedures for concentrating and to some extent, purifying the opium has been known for centuries. In the 1500's a Swiss physician, Parcelsus, prepared a relatively pure extract, laudanum, which is still used today. The isolation and chemical identification of the active ingredients did not occur until the 1800's, when morphine (which comprises about 10% of dried opium powder) was isolated and codeine (which comprises less than 1% of opium powder) was identified. Each of these components is an effective analgesic, and each has substantial potential for abuse. Ironically, when analogues of these compounds were synthesized in the laboratory, one of them, heroin, was hailed as the "hero drug" that could relieve pain without causing addiction!

The narcotic drugs produce an excellent blend of direct pain reduction and the attenuation of the psychological trauma associated with pain. These effects are especially desirable for acute and severe pain associated with injuries, most notably those that occur on the battlefield. As information about the various neurotransmitters and their receptor specificity started to unfold (cf., Chapter 2), researchers began to search for the mechanism of action of the narcotic drugs. The basic question was "Which transmitter substance is mimicked, blocked, or otherwise modified by the opioid drugs?"

The answer, curiously enough, was none of the above. Although specific transmitters (e.g., histamine and serotonin) had been linked to pain, the narcotic drugs did not appear to interact directly with these systems. As neuropharmacological techniques became more sophisticated, it became possible to isolate and identify specific receptors through a procedure that measures receptor binding. The procedure (shown in Fig. 5-3) is complicated, but it can be summarized as follows (see also related discussions of receptor binding in chapters 5 and 6): The compound in question is prepared in radioactive form and injected into an experimental animal. At some later time, usually calculated to coincide with known times for maximum action of the drug, the brain is homogenized and treated in various chemical and physical (e.g., centrifugation) ways until a relatively pure sample of the radioactive compound and its attached cellular components has been formed. This substance is, for all practical purposes, the original drug and the brain's receptors for the drug.

One of the problems with the type of experiment outlined above is that a considerable amount of nonspecific binding can also occur, rendering the results meaningless. Avarim Goldstein's laboratory (cf., Goldstein et al, 1971) had developed procedures that involved special washing of the brain tissue, along with very small amounts of the radioactive drug. Using these procedures and a specific antagonist of opiates, naloxone, Pert and Snyder (1973) were able to show specific binding sites in the brain. Comparison of the relative potencies of various opiate mimickers and blockers showed a close correlation with the ability to bind to these receptors (Fig. 5-4), confirming the notion that these were the receptors that are normally involved in the action of opiate pain relievers (cf., Jaffe & Martin, 1980; Snyder, 1978).

But why should the brain have receptors for an extract of the opium poppy? The only logical answer is that the brain does not have receptors for opium. Rather, the brain must have receptors for compounds produced by the body (endogenous compounds) which happen to share a chemical similarity with the narcotic compounds. Given this conclusion, the race was on to find these chemicals in the body, and to delineate the conditions under which they are released.

In 1975, two groups of investigators (Hughes and Kosterlitz from Scotland and Simantov and Snyder from the United States) independently isolated two substances from pig brain and calf brain that had specific morphine-like properties. Hughes and his associates (1975) dubbed these substances enkephalins, from the Greek meaning "in the head". The substances were small peptides consisting of five amino acids each:

Tyr-Gly-Gly-Phe-Met, and


A decade earlier, Li and associates (1965) had isolated a large pituitary hormone which he called beta-lipotropin (so named because it induces fat metabolism). When the structure of this molecule was shown at a convention, one of Hughes' associates, Howard Morris, was in the audience and made a remarkable observation that can only be likened to recognizing a familiar face in a crowd--he noticed the sequence Try-Gly-Gly-Phe-Met (i.e., met-enkephalin) imbedded in the middle of the long molecule. (See how long it takes you to find it in Figure 7-5 even when you know it is there!) It is suspected that this pituitary hormone may serve as a precursor for at least some of the smaller enkephalins that are formed in the brain.

Later studies have shown that the beta-lipotropin molecule not only contains the met-enkephalin sequence, but several other sequences that are significant to stressful conditions. Positions 4-10 forms the sequence for ACTH, while positions 61-76, 61-77, and 61-91 contain the sequences for alpha-, gamma-, and beta-endorphin, respectively. Because all of these compounds have morphine-like properties, they have been termed endorphins as a contraction for endogenous morphines (Simantov & Snyder, 1976). As indicated in Figure 7-5, the beta-lipotropin is released primarily from the intermediate lobe of the pituitary.

The receptors for these endogenous compounds are located in logically appropriate places. In particular, they tend to be highly concentrated in the limbic system (which is involved with emotional responses), and in the periaqueductal gray area of the brain stem (which is strongly implicated as in pain circuitry).

In summary, there appears to be a system within the brain that can produce opioid compounds and there are specific receptors located in appropriate regions of the brain. We turn now to behavioral experiments that demonstrate the action of these systems.

Pain Reduction Systems

One of the most straightforward schemas of pain reduction systems is that put forth by Watkins and Mayer in 1982. This system, which is summarized in Figure 5-6, involves four types of analgesia: Opioid analgesia from both neural and hormonal sources, and Non-Opioid analgesia from neural and hormonal sources.

The major sources of hormonal opiates arise from the pituitary (especially the intermediate lobe and from the adrenal gland (apparently both the medulla and the cortex of the adrenal). As in the case of other hormones, these opioid compounds are released into the bloodstream and can have their effects on widely dispersed target sites.

There are two major sources of neuronal opiates (i.e., opiates that are released at synapses as neurotransmitters). The arcuate nuclei of the hypothalamus contain a population of cells that are anatomically connected to the limbic system and the periaqueductal gray areas of the brain stem (areas that have been shown to have large numbers of opiate receptors). The second source is an opioid link in the descending periaqueductal gray system. In this case, the opioid transmitter substance inhibits cells that transmit pain signals to the thalamus.

The periaqueductal gray system has a component of fibers that does not utilize an opioid transmitter substance, and there apparently are pain inhibiting compounds (of unknown origin) that are released into the bloodstream as hormones. There is relatively little information about either of these systems, and they are typically discussed in terms of what they are not (i.e., nonopioid) rather than what they are.

The evidence for these four types of analgesia comes from many different experiments, but only a few need be described to show the rationale of these conclusions.

Direct electrical stimulation of the periaqueductal gray system through implanted electrodes produces analgesia. For example, the latency to flick the tail away from a heat source is significantly increased. But what type of analgesia does this represent? The standard test is to determine the response to an opiate antagonist, usually naloxone or naltrexone. If a compound that is known to block the analgesic effects of morphine or opium also blocks the analgesia that is produced by electrical stimulation, then it seems likely that the stimulation is causing the release of endogenous opiates. Naloxone blocks the analgesia produced by periaqueductal gray stimulation. Further evidence for the opioid nature of this effect comes from drug tests that involve morphine. The analgesia produced by electrical stimulation can be enhanced by the administration of morphine. Furthermore, animals that have been rendered tolerant to the effects of morphine show less analgesia with stimulation, and animals that have become tolerant to the effects of electrical stimulation are less responsive to the effects of morphine. Thus, there are three converging results that support the notion that periaqueductal gray stimulation produces analgesia via the release of endogenous opiates:

(a) Naloxone blocks the effect,

(b) Synergism with morphine, and

(c) Cross tolerance with morphine.

It has also been shown that pain itself is a potent stimulus for the production of analgesia. Watkins and Mayer, for example, have shown that electric shock delivered to the front paws of a rat will produce analgesia, as measured by the tail flick response. The administration of naloxone just before the foot shock will abolish this effect. (Naloxone administered after the foot shock does not diminish the analgesia, suggesting that the effect is triggered by endorphins, but not necessarily sustained by them.) The foot shock induced analgesia also shows cross tolerance to morphine, as would be expected from the discussion above.

These naloxone tests confirm the opioid nature of the response, but do not show whether the opiates are neural or hormonal. This distinction is made on the basis of additional experiments. The removal of the pituitary and/or the adrenal gland (the major sources of hormonal opiates) do not diminish the analgesia. Furthermore, transection of the dorsolateral funiculus (the pathway through which the descending periaqueductal gray fibers pass) abolishes the analgesia. Finally, it has been shown that the direct application of naloxone to spinal neurons in the sacral region (i.e., those serving the tail) will prevent the development of analgesia produced by the shock to the front paws. Thus, all of these data converge to suggest that the front paw shock produces analgesia via the small "opioid link" shown in Figure 5-6.

The next aspect of the story seems rather bizarre at first, but the results have been consistent in the hands of Watkins and Mayer. They have shown repeatedly that rear foot shock also produces analgesia, but the effect is not blocked by naloxone. Removal of the pituitary or adrenal is also ineffective, suggesting a neural rather than a hormonal effect. The fibers involved also appear to travel through the dorsolateral funiculus, but from a different origin within the brainstem. These data imply that the pain reduction is neural, but non-opiate in nature.

Prolonged foot shock and/or immobilization produce analgesia that is blocked by naloxone and significantly reduced by removal of the pituitary or adrenal glands (e.g., Lewis et al, 1982). These data support the notion of hormonal opiates that are released by these endocrine glands into the bloodstream.

Although the details of the system are yet to be described, there is evidence for hormonal systems of analgesia that do not involve opioid compounds. Some types of environmental stressors (e.g, cold water swims) can produce an analgesic effect that is not blocked by naloxone, but which requires the integrity of the pituitary and adrenal glands.

These systems of analgesia are peculiar in that different types of pain and different body locations of pain seem to activate different types of analgesia. But why should the rear feet be connected to a different system than the front feet? And why should brief shocks produce one effect, while prolonged shock produces another? These peculiarities may be more apparent than real: The differing procedures may involve differing levels of shock. An attractively simple model of this has been proposed by Forman and Kelsey (personal communication) in a schema that relates the type of analgesia to the impact of the aversive stimulus (some measure of both the intensity and the duration of the painful stimulus). As shown in Figure 5-7, this model suggests that increasing the impact of the painful stimulus can determine which of the four types of analgesia will be produced, with neural opiate analgesia being the easiest to elicit, and hormonal opiate analgesia being elicited only by prolonged or severe pain. This is consistent with much of the experimental literature, including the forepaw/hindpaw phenomenon if one argues that it is easier for the rat to lift the front paws and reduce the impact of the shock (hence, neural opiate rather than neural non-opiate).

A final observation is that all analgesic effects that are based on conditioning (e.g, a reminder of a previous shock can produce analgesia) seem to be based on opiates and can be blocked by naloxone. This appears to be the case even when the original painful stimulus resulted in non-opiate analgesia.

Behavioral Effects on Pain Reduction

Interpretation of pain

The experiments outlined above clearly show that aversive stimulation can produce changes in pain perception. This important influence of the environment on something so fundamental as pain perception came as somewhat of a surprise, but was only the tip of the iceberg in terms of the behavioral interactions that were to be demonstrated. As in the phenomena discussed in the previous chapters, the interpretation of the environment has a profound effect on these systems of analgesia.

The triad design, once again, has been useful in demonstrating the role of behavioral variables in producing analgesia. Although a number of different investigators have used this procedure, a particularly efficient application of the procedure has been developed by Kelsey and his associates (cf., Forman & Kelsey, personal communication). In this procedure, the rats are placed in small restraining cages that have a wheel which can be turned by the front paws. The protruding tail has electrodes attached for delivery of electric shock and can also be positioned in a groove that has a heat lamp for measuring tail-flick latencies in the same session. The typical triad design includes the non-shocked control group, a group that can escape the shock, and a yoked group that receives inescapable shocks. As might be expected from the results discussed in previous contexts, the rats in the first two conditions show no analgesia, whereas the rats that have no control over the shock show a significant increase in the tail flick latencies (see Fig. 5-8).

It should also be noted that the role of the endogenous opiates goes beyond that of mediating the response to direct, painful stimuli from the external environment. A particularly illuminating experiment has shown the effects of social interactions on these systems (Miczek, Thompson & Shuster, 1982). These investigators allowed mice to establish "residence" in their home cages over a period of time. (This involves marking the territory with odors, etc.) They then introduced another mouse into this home territory. Almost invariably, the "intruder" mouse gets attacked and defeated under these conditions, even if the intruder has a "height and weight" advantage. These interactions involve some actual biting of the intruder, as well as a great deal of species specific social postural signals for dominance (on the part of the resident) and submission (on the part of the intruder). The intruder was placed into the resident cage on 10 trials, and allowed to remain each time until 10 bites had been received. As shown in Fig. 5-9, the exposure to defeat led to increasing amounts of analgesia as measured by tail flick latencies. This analgesia declined within an hour after the last session.

Control experiments showed the specificity of this effect by demonstrating that "intruding" into an empty cage had no effect, even if the mouse was "bitten" by forceps while in the cage. Apparently, the social trauma of defeat is an important aspect of this phenomenon.

This effect was shown to be the result of opiate release by the administration of naltrexone, which blocked the development of analgesia. Furthermore, the opiates appear to be playing a role in the central nervous system rather than peripheral effectors, because the administration of a quaternary form of naltrexone, which does not cross the blood brain barrier, did not block the analgesic effect.

Finally, these investigators showed cross tolerance to morphine. In one case, they exposed the animals to the social defeat for 14 successive days, by which time the tail flick latencies had returned to normal (i.e., the mice appeared to be tolerant to the effects of defeat). After this series of defeats, morphine was also ineffective in changing the tail flick response. Turning the procedure around, they implanted time-release morphine pellets that slowly release the analgesic drug over a period of one week. At this time, the mice were placed into resident cages as intruders. The experience of defeat did not change the tail flick latencies. Thus, the cross-tolerance between morphine administration and exposure to defeat were demonstrated (Fig. 5-10).

Placebo and acupuncture

The endogenous opiates also help to explain two types of pain relief that have long been viewed with suspicion: acupuncture and placebo effects. Mayer et al (1976) have shown that acupuncture does produce a reliable decrease in dental pain, and this decrease is blocked by naloxone. Whether this pain reduction is the result of the neural interconnections proposed by the ancient system of acupuncturists or is simply the result of the patients' belief that it will work, remains somewhat problematic.

Although the Western world was reluctant to accept the validity of acupuncture, a related procedure had been used unwittingly in veterinarian practice for many years. The procedure known as twitching involves the firm squeezing of a horse's upper lip in a rope noose. After a few minutes, the horse appears groggy, and can undergo minor surgical procedures with no evidence of pain. This had been interpreted as distracting the horse's attention from the pain of the surgery, but it has recently been shown to increase endorphin levels and the effect can be blocked by naloxone (Lagerweij et al, 1984). It is also interesting to note that long distance runners can sometimes find relief from painful side aches ("stitches") by pinching the upper lip.

The term placebo means "I will please", and has long been used to designate an inactive compound that is administered as though it is an effective drug. In some cases, this is done quite voluntarily, while in other cases, both patient and physician may be misled. In any event, there is now evidence that the endogenous opiates play a role in at least some of the placebo effects. Dental pain is once again the testing ground. In a second series of experiments, Mayer and associates (1976) found that some, but not all, patients showed a reduction in the pain that they experience after they were given medication that was claimed to be a pain reliever. For those patients who responded with a reduction in pain, naloxone blocked the effect and caused an increase in pain. For those who showed no effect of the placebo, the opiate blocker was without effect (Fig. 5-11). Although a lot of additional work needs to be done, it is probably not the case that placebos work for some people but not others. More likely, placebos probably work for virtually everybody, but the conditions under which they are likely to work may differ from individual to individual.

An Overview of the Pain Response

The evidence is clear that pain bears an uneven relationship to the actual physical intensity of the stimulus. When an aversive stimulus is first presented, it almost always produces an immediate behavioral response, and it may even be accurate to view this as a reflexive response. But the processes that change the perception are triggered almost immediately. The exposure to the painful stimulus triggers feedback mechanisms that dull the intensity of the pain. The ability or lack of ability to control the painful stimulus further modulates this perceptual change. In the next section of this chapter, it will be shown that the experience and interpretation of painful events not only changes the perception of pain, but also has far reaching ramifications in terms of the general ability of the organism to respond to stressful challenges.


Survey of the Immune System

The biological self.

The concept of self has been important in several different contexts. In both philosophy and psychology, the notion of the individual being as a unique entity has been pivotal in writings and theoretical developments. The term is so entrenched in our language, that we rarely notice the philosophical implications when we say, for example, "I'd rather do it myself." Yet, a failure to distinguish the difference between self and others adequately or early enough can lead to profound disturbances that are likely to be labeled psychotic.

If the self really is a unique entity that makes up an individual's being, then it is vital to maintain that entity. This requires two fundamental abilities: The individual must be able to discriminate the boundaries between self and nonself, and must be able to defend against forces that would blur this distinction.

The biological self would appear, at first glance, to be so obvious that the concept would be useless. Physical boundaries alone seem to make the distinction between the self and nonself even plainer than the nose on one's face. The biological self became important when sexual reproduction was "invented" and a unique set of information about one self (or is it one's self?) was merged with the information about another self to create a third self. This new importance of individuality was a benchmark in evolutionary history, but the evolutionary advantages also introduced more stringent requirements to be able to recognize and defend the self against the nonself. The immune system is the first line of defense in this task.

The field of immunology is currently one of the most complicated and most active areas of biological research. The increasing knowledge about this system has all but eliminated several major scourges of mankind, such as measles, smallpox, polio and diphtheria. We can protect our selves by taking the appropriate "shots". The importance of this area to behavioral pharmacology is that the environment and the interpretation of the environment have major effects on the immune system. This has led Ader and others (e.g., Ader, 1981) to coin the term psychoneuroimmunology. We turn now to a cursory treatment of the mechanics of the immune system, to be followed by a discussion of the ways in which behavior can place its mark upon this system to either fortify or break down the defenses of our biological selves.

The normal role of the immune system is to recognize foreign (i.e., nonself) substances and create a locally hostile environment that will eliminate them from the system. This can occur in response to the introduction of disease producing bacteria or viruses, tumor cells, parasites, transplanted organs, inappropriately matched blood transfusions, and a host of other substances. In this, its normal role, the immune system is somewhat of a silent warrior, and does not command the attention of the individual. However, for a sizeable proportion of the population, the immune system becomes very obvious by virtue of its somewhat inappropriate response to substances that pose no real threat to the individual. These are the individuals who suffer from allergies to such things as pollen, cat dander, milk, and jewelry, to name a few. Even more serious than over responding to a harmless foreign substance is a failure to recognize self as self. This can occur in a variety of autoimmune diseases such as arthritis, Parkinson's disease, myasthenia gravis, some forms of diabetes, and others (the list is growing).

There are two major ways in which the immune system can respond to a challenge: One of these is a humoral response which can occur almost immediately in a sensitized individual. This involves substances that were created by the immune system and which circulate in the bloodstream until they come into contact with the specific foreign body. The other is a cellular response that involves the proliferation of special blood cells that can react to the specific foreign substance. This reaction is referred to as the delayed response, because it typically requires about 48 hours to develop. The two categories of immune response are mediated by specialized leukocytes (white blood cells) that arise from nonspecialized stem cells formed in the bone marrow (B-cells) and in the thymus gland (T-cells). Refer to Fig. 5-12 for a schematic diagram of the reactions that are being outlined.

Humoral responses.

The humoral response is initiated when a B-lymphocyte recognizes the presence of a foreign substance, the antigen (refer to Fig. 5-13; after Buisseret, 1982). The magnitude of this response is partially determined by the effects of T-lymphocytes, which are termed helper or suppressor T-cells, depending on their role in the interaction. If this initial encounter is interpreted as a challenge from a foreign body, their is a tremendous increase in the metabolic activity of the B-cell and it begins to manufacture and release antibodies that are chemically specific to the original antigen. These antibodies are the immunoglobulins, of which there are five types and several subtypes. One of these, Immunoglobulin-G, has been strongly linked to populations that are exposed to parasitic worms. This same form appears to be involved in allergic responses, presumably because the suppressor T-cells in some individuals allows the initial reaction between antigen and B-cell to continue.

The immunoglobulins that are released by the B-cells interact with mast cells (so-named because they appeared to the German investigator to be "stuffed"), large cells that are found in connective tissues, in the membranes of the intestines, eyes, and respiratory tract, in the skin, and in the lymph glands. These cells contain numerous granules within their cytoplasm, and their surface membrane has several hundred thousand receptors for antibodies (i.e., immunoglobulins). These receptors are nonspecific, in that they will bind any form of immunoglobulin to the surface of the mast cell. However, when two specific antibodies occupy adjacent receptors, the pair forms a highly specific receptor for the original antigen. This sort of "piggy-back" arrangement is a highly efficient way for a single population of cells, the mast cells, to develop unique sensitivities to any of thousands of potential antigens that might be encountered. (This type of arrangement has not been described for neurons, but it is such a clever mechanism that it would be surprising if it were only used in this one system.) The number of mast cells that have this sensitivity conferred upon them is roughly related to the amount of immunoglobulin formed by the B-cells in their first encounter with the antigen.

Once the mast cells have been sensitized, the system is ready for an immediate response when the original antigen is encountered again. When the receptor pair identifies the antigen, calcium enters the mast cell and the granules move to the periphery and are released in a manner that is comparable to the release of a neurotransmitter substance. These granules are comprised of a group of compounds (histamine, serotonin, heparin, blood platelet activators, etc.), which are collectively termed the preformed chemical mediators. These chemicals have widespread effects on blood vessels, respiratory membranes, smooth muscles, and the blood itself, producing what is commonly called the hay fever reaction. At the same time, there is also an increase in the production of prostaglandins (which have complicated effects on respiratory membranes, mucous secretions, blood clotting factors, etc.) and the leukotrienes. The leukotrienes are many times more potent than histamine, and are responsible for the asthmatic symptoms of contracted airways, dilated and leaky small blood vessels, and painful, itchy nerve endings. In severe cases of allergy, this set of reactions can lead to what is called anaphylactic shock, including the danger of respiratory collapse and death: Nature's way of telling us that we have encountered a foreign substance.

It should be noted at this point, that virtually all drugs (and foods for that matter) are foreign substances, and might be expected to induce an allergic sensitization. Although some drugs can sensitize an individual and lead to an anaphylactic response (penicillin is one of the most common examples), these reactions are rather uncommon. The reason for this is that the immune system is designed to respond to large, nonself proteins. The sensitization process (i.e., the development of specific receptors) is based upon some small part of the surface of the antigen molecule, which is termed the epitope. The presentation of this epitope alone is usually ineffective. Most drug molecules are small relative to proteins, and even if bound to a protein after administration, the likelihood of sensitization is rather slim.

Cellular responses.

The cellular immune response involves the sensitization and proliferation of T-leukocytes (see Fig. 5-14), which mature in the thymus gland. When an antigen is encountered, it is bound to a special macrophage which presents it to the T-cell. The T-cell receptors represent what has been termed the major histocompatibility complex (MHC), which in some sense, can be viewed as the "self-template" against which potentially nonself substances can be compared. The recognition of a foreign substance triggers the release of interleukin-2 which stimulates cell division in the T-cells and produces a substance called gamma-interferon which stimulates prostaglandin release, causes fever (which also promotes cell division), and stimulates both the MHC receptors and the presenting by macrophages. This positive feedback system causes a marked proliferation of sensitized T-cells that respond to the specific antigen. These sensitized T-cells release a group of compounds called lymphokines that produce a characteristic set of effects: (a) dilation of small blood vessels resulting in local redness and warmth, (b) leaky blood vessel walls resulting in swelling, and (c) congregation of macrophages, including phagocytes for removal of foreign substances, damaged tissues, etc. This proliferation of T-cells and the associated macrophages accounts for the increase in the white blood cell count that characterizes the response to infections and injuries.

The immune system obviously is not a static system. It can respond to any of host of potentially threatening molecules in an efficient and specific manner. However, the likelihood that a particular molecule will trigger a reaction is influenced by both genetic and environmental factors. Buisseret (e.g., 1982) has studied this rather extensively in the case of allergies to milk. The genetic link is strong, though not complete: If both parents are allergic, the offspring have a 58% chance of being allergic. If one parent has the milk allergy, the offspring has a 38% chance of being allergic. The rate is only about 12% for those who do not have a family history of milk allergies.

Early environmental factors also play an important role. Buisseret (1976) cites an early study by Grulee (1943) which showed that 36% of breast fed babies contracted an infection of some sort and .0.13% died as a result. By contrast, babies that were bottle fed were almost twice as likely to contract an infection (63%) and far more likely to die as a result (7.56%)! This could be attributed, in part, to more sanitary conditions surrounding breast feeding, but the bulk of the effect is probably due to the transfer of immunities via the mother's milk (cf., Appleton & McGregor, 1984). There is also some evidence that breast milk may help to prevent proteins from crossing through the intestinal linings where they can sensitize the immune system. Buisseret (1978) followed up these results with an investigation of the interaction between type of early feeding, genetics, and the likelihood of showing an allergy to milk. As shown in Figure 5-15, the breast fed offspring of parents who do not have milk allergies have virtually no chance of developing the allergy. Bottle fed offspring of parents who have allergies show a 60% likelihood of having the allergy. Clearly, the immune system can be influenced by both genetics and the early environment. Given this degree of flexibility, it should not be too surprising that later behavioral influences can also be demonstrated. We turn now to these effects.

Behavioral Effects on the Immune System

Interpretation of the environment.

When Selye (1956) began his pioneering work in defining stress, he noted that general trauma could result in a decrease in the size of the thymus gland and spleen, as well as a general lymphopenia (decrease in number of leukocytes in blood). Although these and similar effects had been related to the body's lowered ability to respond to an immunological challenge, the degree of flexibility in the immune system and the importance of behavioral factors has only been realized in recent years. For example, one of the more dramatic studies along these lines demonstrated that placing rats on a turntable for a brief period each day dramatically increased the growth of implanted tumors (Sklar and Anisman, 1979). The importance of this finding is that a relatively mild stressor that is not directly related to a specific disease (in this case, the tumor) can greatly diminish the ability of the organism to defend against the disease, once it is encountered.

This early work on resistance to tumors has been sharpened considerably by studies that have looked more closely at the behavioral components. Visintainer and associates (1982) implanted a suspension of tumor cells into the flanks of rats, waited 24 hours, then exposed the rats to a set of electric shocks in the familiar triad design. A control group received no shock, a second group received 60 escapable shocks delivered on a variable interval schedule, and a third group received the same schedule of shocks, but the shocks were inescapable. Only 27% of the rats that received inescapable shocks rejected the tumor, whereas more than half of the control group (54%) and the rats that received escapable shock (63%) rejected the tumors. The rats in the two shock groups received exactly the same number and duration of electric shocks, but the lack of behavioral control over shock termination doubled the likelihood that the tumor would get out of control and kill the rat!

The ability to reject a tumor presumably requires the ability to launch a cellular immune response. This ability has been assessed more directly in studies that have used T-cell proliferation as the measure of the viability of the immune system. Laudenslager and associates (1983) used the triad design of shock administration during a single 80-minute session. At the end of the session, the took blood from the animals, extracted the leukocytes, and treated the leukocytes with T-cell mitogens (either concanavalin A or phytohemagglutinin, abbreviated CON-A and PHA, respectively). These mitogens stimulate cell division, but the amount of cell proliferation depends upon the tendency of the cells to proliferate before they were removed from the system. These investigators demonstrated that exposure to inescapable shock greatly reduced the amount of T-cell proliferation, whereas the exposure to the same amount of escapable shock had no effect (Fig. 5-16). These results demonstrate two important effects: The lack of a coping response suppresses the immune system's ability to respond, and this suppression is triggered very quickly (recall that the blood samples were taken immediately at the end of the session), even though its effects might not become manifest for days or weeks later (e.g., in the case of a challenge by tumor cells).

Learned immune responses.

There is also evidence that changes in the immune system can be learned. Ader and Cohen

(1982) demonstrated this possibility using a strain of New Zealand mice that are genetically predisposed to suffer from serum lupus erythematosus (SLE), an autoimmune disease. This disease is used as a model for similar disorders in humans, and involves a breakdown of connective tissue and a variety of secondary symptoms such as kidney failure. These mice have a relatively short lifespan which can be prolonged by treatment with cyclophosphamide, a drug that suppresses the immune system. The two groups that were used for comparison were the untreated control group (25% of the mice die of their affliction by the time they are 10 weeks old) and a group that received a standard 8-week regimen of chemotherapy (25% mortality is not reached until about 35 weeks). If the animals receive only 4 injections (during alternate weeks of the 8-week treatment period), the chemotherapy is considerably less effective, and the 25% mortality figure is reached at about 20 weeks. Ader and Cohen used this intermediate level of treatment to test for the possibility of conditioning. Prior to each drug injection, the mice received a distinctively flavored saccharin solution. On the intervening weeks, they received the saccharin solution followed by a placebo injection of saline. The question was, would the immune system be suppressed because of the prior association of the distinctive flavor with the injection of cyclophosphamide? The answer was yes: The rats that received these conditioning trials did not reach the 25% mortality figure until 25 weeks, about five weeks longer than the group that simply received the staggered drug injections.

MacQueen and associates (1989) used a somewhat different Pavlovian conditioning approach to demonstrate the learned release of preformed mediators from mast cells. In this experiment, they paired an audiovisual cue with the injection of egg albumin (an antigen that promotes mast cell activity) into rats. Later presentation of the audiovisual cue alone stimulated the release of the preformed mediators to the same degree as re-exposure to the antigen itself.

The studies that demonstrate changes in the immune system as a result of learning have important implications for medical treatment. Particularly in the case of chemotherapy for cancer, the drugs such as cyclophosphamide are highly toxic, and conditioning procedures might be able to greatly reduce the amount of drug that needs to be administered. Similarly, a long list of ailments including colitis, irritable bowel, asthma, and various food allergies may have a considerable learned component which can be treated more effectively by behavioral means than by drugs. Another intriguing possibility suggests that learning can confer a degree of immunity against the changes in the immune system. For example, McGrath and Kelsey (personal communication) have preliminary data showing that prior exposure to escapable shock can protect rats from the suppression of immunity that normally results from shock that cannot be escaped.

Interaction with endorphins.

The similarity of the behavioral treatments that suppress the immune system and those that trigger the release of endorphins is probably more than a chance occurrence. Shavit and colleagues (1984) have demonstrated the presence of an opioid link in immunological suppression. In particular, they found that a regimen of foot shock that released opioid compounds also suppressed the number of natural killer cells (rather nonspecific immune cells) from the spleen. Both effects were blockable with naloxone. By contrast, a regimen of foot shock that resulted in nonopioid analgesia had no effect on the immune response and was not blockable by naloxone. Further evidence of this opioid link was suggested by the finding that large dosages of morphine also suppress the immune response.

Implications for receptor function.

These changes in the immune system are particularly important in the area of behavioral pharmacology because of their potential for changing the viability of transmitter systems. This has been most clearly demonstrated in the case of myasthenia gravis. This disorder involves the progressive decline in the function of somatic muscles characterized, as the name implies, by a grave weakening of the muscles. Over the years, some researchers have attributed the disorder to a deficiency of acetylcholine (the neuromuscular transmitter) while others have argued that the disease involves a deficiency in the number or sensitivity of the receptors. Either argument is consistent with the observation that patients benefit from treatment with anticholinesterase compounds.

Myasthenia gravis has been strongly linked to an autoimmune disorder which attacks the receptors for acetylcholine. In 1973, Patrick and Lindstrom extracted nicotinic receptors from electric eels and injected a suspension of these receptors into rabbits. The immune systems of the rabbits recognized the nonself nature of these proteins, and developed antibodies. However, the antibodies were not sufficiently specific, and reacted not only to the nicotinic receptors of the eel, but also to their own nicotinic receptors. As a result, they developed experimental allergic myasthenia gravis (see Fig. 5-17).

It is one thing to show that myasthenia gravis can be mimicked by manipulation of the immune system, but this does not necessarily mean that this is the normal cause in humans. However, the evidence was soon to follow. Almon and associates (1974) found antireceptor antibodies in 87% of patients suffering from myasthenia gravis. As a result of these antibodies, there was a decline in acetylcholine receptor activity of 70-90 percent. For some reason, these patients have developed antibodies to their own receptor sites. Unfortunately, knowing the cause has not provided the cure. Corticosteroids inhibit the immune response (cf., Fig. 5-14), but such treatment has a host of side effects, many of which are dangerous.

The specific chemical relationships between transmitters, receptor sites, and antibodies provide a generously complicated number of ways in which the immune system can cause behavioral dysfunction. Consider, for example, an experiment by Shechter and associates (1984) which was investigating insulin and insulin receptors, but could be equally relevant to a neurotransmitter system (see Fig. 5-18). They began the series by immunizing mice with insulin from cows or pigs. The mice developed specific antibodies (called idiotypes) to this foreign substance. These idiotypes showed some of the characteristics of insulin, and also triggered the formation of anti-idiotypes which were, in effect, antibodies against the mice's own insulin receptors. As a result, the mice developed symptoms of diabetes.

Autoimmunity and behavioral disorders.

What does all of this mean with respect to behavioral disorders? It is unlikely that we are going to encounter suspensions of eel receptors, and we can probably rest assured that we will not be exposed, for example, to the antibodies that mice have formed against the receptors for one of our own hormones. The point of these experiments is that the machinery is there, and it can easily be tricked into developing immune responses that can interfere with the normal process of receptor activity. Thus, immune disorders (or more specifically, autoimmune disorders) become primary candidates for the cause of not only myasthenia gravis, but perhaps schizophrenia, Parkinson's disease, multiple sclerosis, depression, diabetes, and others. The evidence is already beginning to accumulate for several of these.

An unexpected link between hormones, behavior and autoimmunity has been put forth by Geschwind & Behan (1982). Geschwind formed two groups of subjects who had been identified for extreme handedness. Those in the left handed group showed a 12-fold increase in learning disabilities, but also had a disproportionately high number of artists, musicians, engineers, and mathematicians (skills that have been linked to right hemispheric functions). Curiously, 11% of these individuals had some sort of autoimmune disease, whereas only 4% of the right handers suffered from these disorders. The intriguing link to hormones comes from several different observations. For reasons that are unknown, testosterone inhibits the growth of the left hemisphere and of the thymus gland (a major organ of the immune system). The same region of the chromosome that determines the major histocompatibility complex (the receptor that determines self vs. nonself) may also influence the weight of the testes, serum testosterone levels, and the sensitivity of the receptors to testosterone.

At the present time, many of these experiments and observations offer little more than hints at possible effects. But the results are tantalizing, and it is very likely that within a few years it will become increasingly clear that the immune system is integrally involved in the causes of many behavioral disorders, the response to drugs, and that behavior can, in turn, modify these interactions.


An attractively simple view of pain and other aversive events is that they are always bad and that the body always mobilizes its energies to either eliminate the aversive stimulus or to escape from the situation. This simple view is wrong. We have seen in Chapters 4 and 5 that painful stimulus per se are either inconsequential or (perhaps) beneficial. The harmful aspects of aversive stimuli inhere not in the stimuli but in the lack of an appropriate coping response. The ability to predict when an aversive stimulus sill occur is good; the ability to actually control it is better; the ability to predict and control is even better. In the absence of these, the organism can suffer a variety of consequences including ulcers, the release of endorphins, the suppression of the immune system, or even sudden death.

The different responses that occur when the opportunity for coping verses no coping prevails is paralleled by the general anatomical and physiological features of the brain regions that mediate emotional responding (see Figure 5-19). The structures of the limbic system appear to be primarily responsible for analyzing the emotional tenor of our environment. The structures of the limbic system receive information from both the outside world and the body, analyze this information, and contribute to the body's response to the situation via messages to the hypothalamus and pituitary. These structures are involved in many aspects of behavior, but it is instructive, in the present context, to analyze the types of responses that are mediated by the anterior and posterior portions of these two structures.

The posterior regions of the hypothalamus and pituitary are responsible for sympathetic arousal, including the stimulation of the adrenal medulla (to release E, NE, DA, and endorphins) and the adrenal cortex (to release mineral corticoids and stimulate inflammatory responses). There is a tendency to view these responses in a negative light because of the nature of the situations that lead to these responses, but in fact, these are energizing responses that improve the organism's's ability to interact with its environment. Appropriately, these responses occur when the environment affords the opportunity for something effective to be done-- for example, fighting, fleeing, or coping.

The anterior portions are normally involved with a variety of constructive functions: adjustments of the vegetative responses of the parasympathetic system, production and regulation of a variety of hormonal systems, and so forth. But when the organism confronts an aversive situation for which there is no obvious coping response, these anterior regions overreact and produce a physiological environment that causes tissue damage or otherwise disrupts bodily functions. The outflow of the parasympathetic system erodes the ability to mount a physical response, the adrenal gland releases glucocorticoids which, in high concentrations, can cause tissue damage. The release of endorphins dampens the organism's ability to monitor the environment and may contribute to the suppression of the immune system.

In 1932, Cannon wrote a book entitled The Wisdom of the Body in which he repeatedly demonstrated the adaptive and appropriate responses of the body to such things as hunger, thirst, exercise, danger, and so forth. His arguments were and remain convincing-- the accuracy and complexity of the body's responses is nothing less than awe-inspiring. It is difficult, however, to see the wisdom of some of these responses to situations that do not offer prediction or control. One might argue that the body (like the proverbial customer) is always right, and that we do not see the situation with sufficient clarity to recognize the benefits. Alternatively, one might argue that the ability to interact effectively with the environment is so crucial to the survival of complex organism that the system fails to function when these conditions do not prevail. (This latter view will receive further support in the next chapter.)

The diversity of the behavioral and physiological responses has complicated and enriched the ways in which we view drug effects. The phenothiazines, the benzodiazepines, and even the beta adrenergic blockers can each decrease the emotional response, but the mechanisms make more sense when viewed within the appropriate behavioral and neurochemical context. Opium and related compounds are not just blocking pain, they are interacting with a multifaceted system that uses the body's own opiates. The important interactions of the immune system with behavior implies that any drug that changes these types of behavioral responses will also be likely to have indirect effects on the immune system. The next two chapters extend the discussion of these issues through an analysis of two different types of mental illness.



1. Separate anatomical pathways in the spinal cord carry acute, sharp pain in fast fibers and more chronic, dull pain in slower fibers.

2. Pain thresholds have been determined by several methods, including the flinch/jump test, the paw lick test, and the tail flick test.

3. Opium, an extract of the poppy plant, contains both morphine and codeine; heroin is a synthetic analogue.

4. Specific receptors for opiate drugs have been found in the brain. These receptors mediate the effects of endorphins, a group of endogenous compounds that have morphine-like effects.

5. A large pituitary hormone, B-lipotropin, contains the amino acid sequences for several smaller peptides that are involved in the stress response.

6. There are four systems of pain reduction: opioid from neural and hormonal sources, and nonopioid from neural and hormonal sources.

7. Exposure to inescapable pain or social defeat results in analgesia. In many cases, depending on the precise environmental conditions, this analgesia can be blocked by opiate blockers and is cross tolerant with morphine.

8. Placebo effects and acupuncture appear to be mediated by endorphins.

9. The immune system is involved with recognizing and defending the biological self.

10. The immune system has two major modes of responding: a humoral response that involves circulating immunoglobulins, and a cellular response that involves the proliferation of T-cells.

11. Relatively mild stressors, if not controllable by the individual, can lead to suppression of the immune system.

12. The failure of the immune system in various ways can increase the vulnerability to diseases, trigger allergies, or lead to autoimmune disorders.

13. Immune suppression can be assessed indirectly y measuring the susceptibility to tumor cells or disease, or it can be measuring directly by determining the amount of T-cell proliferation that results from treating a blood sample with mitogens.

14. Both Parkinson's disease and some forms of diabetes seem to involve an autoimmune response to one's own receptors-- a finding that may have important implications for a variety of behavioral disorders.

15. Testosterone appears to inhibit growth of the left hemisphere and the thymus gland, a finding which may account for the abnormally high incidence of autoimmune disorders in left handed males.

16. The posterior portions of the hypothalamus and pituitary mediate the sympathetic arousal response; the anterior portions mediate the overreaction of the parasympathetic system,the release of endogenous opiates, and other responses that accompany the inability to cope.






Anaphylactic shock




Arcuate nucleus

Autoimmune diseases

B cell


Bright pain

Cellular response





Flinch/jump test


Humoral response



Limbic system



Mast cell




Myasthenia gravis




Paw-lick test

Periaqueductal gray



Preformed chemical mediators


Receptor binding

Receptor antibodies


Social defeat

Substance P


Tail-flick test

Triad design