r-

Medical Hypotheses

The Immune System is a Key Factor in the Etiology

of Psychosocial Disease
R. S. SMITH
4 7 18 Meridian Ave., Suite 208, San Jose, CA 95 118, USA

Abstract - The immune system is proposed as the key to understanding the etiology and
treatment of psychosocial disease. There is a dense communication network between the
immune system and the central nervous system (CNS). Immune cell cytokines, via direct
action on the CNS, induce fever, alter sleep, pain perception and pituitary hormone release
and reduce appetitie and activity in animals. Interleukin-2 and interferon given to humans
result in global behavioral and cognitive pathology. Activation of the immune system by
pathogens produces global cognitive and behavioral pathology also. Recently, controlled
trials have demonstrated that diet can cause psychosocial disease, presumably by an immune
mechanism. Immune system abnormalities have been identified in manic-depressive psychosis,
schizophrenia and alcoholism. Lithium carbonate is not only prophylactic for all three of
these diseases, but it also powerfully stimulates the immune system. This is proposed
as the mechanism of lithiums therapeutic effect. The antipsychotics, haloperidol and the
phenothiazines, affect the immune system also. The rapid development of AIDS dementia
complex can be explained by the remarkable influence the immune system has on the CNS.

Introduction

secret agent appears to be the immune system. Arterial macrophages ingesting altered low density
lipoprotein is probably the key event in the development of lipid filled fibrous plaque (1).
Psychosocial disease has not participated in
the triumphant advances of molecular biology
and modern medicine. The term psychosocial disease covers a broad spectrum of maladaptions
and malfunctionings of the CNS, including mental illness, addictions, criminal behavior, suicide,
violence, educational and employment failure,
poverty, eating disorders, dementias and child

Elucidating the role of the immune system in hu-

man disease has always resulted in striking advances in understanding, preventing or treating illness. The relationship was shown very early for
infectious disease, but for chronic degenerative
pathologies like atherosclerosis and cancer, insights into the relationship had to wait for basic advances in biomedical science. For example, after 100 years of study, the pathogenesis of
atherosclerosis is now yielding its secret and the
Date received 8 February 1990
Date accepted 27 March 1990
49

50

abuse. It remains the only major disease entity

where there has been little overall progress in theory or practice. In fact, psychosocial disease for
the most part is a growing and explosive epidemic
with no rational view of a solution in the near
or distant future. Furthermore, it is the only disease where immune system involvement has not
been generally incorporated into understanding its
pathogenesis.
I propose the following medical hypothesis:
the immune system has an influence, sometimes
quite distinct and powerful, on the CNS. The immune system plays an important, often times critical causal role in many psychosocial disease syndromes. There will be a significant leap forward
in our understanding, treatment and prevention
of psychosocial disease when the immune system
involvement is elucidated. Breakthrough progress
will not occur in the prevention of treatment of
psychosocial disease until the immune system involvement is understood.
There are five major areas of support for this
hypothesis: 1. the effects of cytokines on the CNS;
2. the universal experience of mental pathology
accompanying activation of the immune system;
3. the rapid development of a dementia complex
in AIDS patients; 4. the extensive, but neglected,
literature on food allergy as a cause of behavior pathology. The relationship between diet and
behavior pathology has now been confirmed in
three controlled trials; and 5. the effect of lithium
and other neuroleptics on the immune system.
It is proposed here that lithiums activation of
the immune system will be found to be the long
sought after explanation for lithiums therapeutic
and prophylactic effect on manic-depressive psychosis, schizophrenia and alcoholism.
The effects of cytokines on the CNS
Seventeen protein cell regulatory substances, called
cytokines, produced by various immune cells,
have been identified (2). They have powerful and
multiple overlapping regulatory actions on a wide
variety of cell types. The immune cells themselves
are important targets of the cytokines, meaning
the immune system is, to a great extent, self regulatory.
There are many factors, in addition to cytokines, which appear to regulate immune cells.
For example, receptors for various secretions of
the central and peripheral nervous system have
been found on immune cells. In 1979 opioid re-

MEDICAL HYPOTHESES

ceptors on human T-lymphocytes were discovered. The CNS produced opioid, methionineenkephalin, binds with those receptors and results in proliferation of T-lymphocytes (3). In contrast, the opioid alpha-endorphin has inhibitory
effects on T-lymphocytes (4). A recent review (5)
with 84 references covers the extensive effects of
acetylcholine on lymphocytes. In addition, somatostatin, vasoactive intestinal peptide, peptide
histidine isoleucine, peptide histidine methionine,
and glucocortical hormones restrain immune cells
whereas dopamine and substance P stimulate immune cells (6, 7, 8). Plats-Salaman (9) has compiled a list of 29 CNS produced neuroregulators that affect immune cell functions, which include the above plus gonadotrophin-releasing
hormone, growth hormine, prolactin, gonadotrophic
hormones, ACTH, thyrotropin stimulating hormone, melanocyte stimulating hormone, vasopressin, oxytocin, insulin, bombesin, and calcitonin. CNS influence on the immune system is extensive, complex and well documented.
The CNS-immune system axis is not a one-way
street. Like ail other regulatory networks, this one
has a feedback circuit; that is, immune cell secretions influence the brain. The significance of
the feedback circuit has not been well appreciated. It is primarily viewed as a mechanism for
the brain to have more accurate and effective control over the immune system. The possibility that
immune cells could have a powerful influence on
the functioning of the CNS, with the organismic
consequences being altered behavior, has been neglected.
Six cytokines, namely, interferons alpha, beta
and gamma, tumor necrosis factor (TNF), interleukin-1 (IL-l), and interleukin-2 (IL-2), induce
fever via direct access to the temperature regulating center in the hypothalamus (2, 9). IL-1 also
promotes sleep, reduces perception of pain, induces the pituitary to release hormones, regulates
nerve growth factor synthesis, suprersses appetite
and reduces locomotion by acting directly on receptors in the hypothalamus and other areas of the
CNS (9).
Having access to the hypothalamus is of considerable practical and theoretical importance, since
the hypothalamus is the premier regulatory center
of the brain. The endocrine system, with all its
long and short term effects on behavior, is controlled there. Pain, pleasure, rage, sleep, hunger,
satiety, addictions, and to a great extent, learning
are controlled at this remarkable center (10, 11).

THE IMMUNE SYSTEM IN THE ETIOLOGY

OF PSYCHOSOCIAL

In addition, the hypothalamus also appears to be

a master control center for emotion (12). If any
area of the brain could be called the psychosocial
disease center, it would be the hypothalamus. A
primary CNS regulatory target of cytokines is the
hypothalamus.
TNF receptors have been found in the hypothalamus, brain stem and other CNS areas. TNF
increases hormone release from the hypothalamus
and suppresses appetite by direct action there also.
IL-2 receptors have been found in brain. IL-2 enhances hypothalamus activity, inhibits food intake
and acetylcholine release from various CNS areas
and stimulates sympathetic nerve cell branching.
Food intake is also suppressed, via a direct effect on the CNS, by IL-6, B-cell growth factor,
tuftsin, platelet activating factor and interferons
alpha, beta and gamma (9).
A great deal of work has been done demonstrating immune cells, in addition to cytokines, produce a wide variety of neuroendocrine peptides.
Lymphocytes manufacture ACTH, beta-endorphin, corticotropin-releasing
factor, thyrotropin
stimulating hormone, and human chorionic gonadotropin. Monocytes produce somatostatin like
peptides, bombesin, substance P, beta-endorphin,
and ACTH (6). These immune system products,
which mimic neurohormones in structure and action, provide additional ways for immune cells to
regulate the CNS.
The presence of immune cells in the CNS
further enriches the complex web of CNS-immune system interactions. T-lymphocytes can pass
through the blood-brain barrier and enter the
CNS. Microglia, derived from blood monocytes,
are transformed into macrophages when activated
(9). Cytokines and neuroendocrine peptides produced by immune cells outside the brain not only
affect CNS control centers but also the microglia
and T-lymphocytes residing in the brain. These
CNS residing immune cells in turn will produce
their own cytokines and neurohormones which
will directly affect the brain.
We see that there is an exceedingly powerful
and complex two-way relationship between the
immune system and the CNS. The vast majority of researchers have conceptualized the CNSs
influence over the immune system as the dominant and medically important part of the relationship. The hypothesis of this paper emphasizes the
other direction, that is, the immune systems influence and control over the CNS as a dominant
and medically important part.

DISEASE

51

Cytokines administered to human subjects. Cytokines have been given to cancer patients, and
the consequences are global behavioral and cognitive pathology. In a detailed longitudinal study
(13) of 44 metastatic cancer patients given IL-2
and lymphokine-activated killer cells, the majority
experienced profound neuropsychiatric changes.
The severe CNS changes had a 4 - 6 day latency
period before appearance, were dose dependent
and returned to baseline after treatment was discontinued. 42% of the high dose patients had
acute and severe cognitive-behavioral
changes:
delirium, disorientation in time, space and person, delusions, paranoia, irritability, incoherence,
memory loss, inability to concentrate, depression,
agitation, severe fatigue, anorexia and malaise.
Many required physical restraints and neuroleptic
medication. A total of 71% of the high dose patients had either moderate or severe changes. Only
32% of the low dose patients had moderate to
severe changes. The authors stated the profound
CNS effects were dose limiting for the treatment.
There were no significant personal factors
found that were predictive of the severity of individual CNS response. There were premonitory
symptoms occurring during the first days of treatment which invariably preceeded the more severe
changes. The premonitory symptoms were severe
lethargy, impaired memory, vague responses, slow
responses, illogical and incoherent responses, impaired attention, irritability, mild aphasia, paraphasis, intermittent disorientation, disconnecting
medical equipment, and insistence on leaving (14).
Interferons have also been used on cancer patients with profound CNS consequences. Severe
fatigue was most common and sometimes persisted up to 4 weeks post-treatment. With repetitive doses patients became profoundly and increasingly lethargic, with or without accompanying confusion. Somnolence, decrease of mental status, mood changes (depression), dysphasia, loss of smell and taste, hallucinations and
disorientation in time and place were observed.
(15) In a study of 11 patients, the authors wrote,
All patients became pyrexial and complained of
anorexia, fatigue and general malaise.. . . further
questioning revealed them to be withdrawn, slow
to answer questions and totally disinterested in
their surroundings, sleeping for most of the day
(16). In another study of 10 cancer patients using interferon, all patients appeared duller, inattentive and disinterested within the first week.
Eight patients experienced psychomotor retarda-

52
tion with unequivocal impoverishment of spontaneous movements, gestures, expressions, speech
and thought. They displayed varying degrees
of bradykinesia and social withdrawal consonant with their complaints of decreased energy
and signs of a disinclination to act or think.
. . .They frequently ignored eating, various aspects
of grooming and regular daily activities. Their disinclination to act was rationalized by the patients
as loss of appetite and lack of energy. All patients took daily naps (17). Several paradoxical
changes were noted: 8 patients with significant
anxiety before therapy did not report anxiety after
one month of interferon; likewise suicidal ideation
was eliminated in 4 out of 4 patients, and episodic
depression was reduced in 3 out of 5.
The experiments with IL-2 and interferon provides clear, convincing evidence that cytokines
have an astonishing effect on the CNS. To date,
17 cytokines have been discovered, which means
we still have 15 more to test on humans for behavioral effects. If the first two are a clue to the
other 15 then there remains a rich vein to explore
for immunological causes of psychosocial disease.

Symptoms of infectious disease

Probably everyone has experienced the malaise,
fatigue, depression, withdrawal, lack of motivation, reduced appetite, and mental lethargy that
accompanies infection and fever. Delirium can result if the infection is severe enough (18). These
universal symptoms of infectious disease clearly
illustrate that activation of the immune system
predictably and massively affects CNS function.
An individual with significant infectious disease is
usually temporarily sociahy and mentally dysfunctional. Delirium is difficult to distinguish from
psychotic illness. The work with IL-2 and interferon on cancer patients show the temporary mental and emotional dysfunction accompanying infection is most likely caused by the cytokines of
the activated immune system.
The significance of the mental illness accompanying infectious disease has not been appreciated,
primarily because it is temporary. If it were permanent, it would be indistinguishable from typical mental illness and psychosocial disease. Using IL-2 or interferon, we now have the ability
in principle to experimentally produce permanent
mental illness in normal human volunteers. This
possibility should illustrate that the symptoms of
infectious disease are a powerful model for mental

MEDICAL HYPOTHESES

illness and psychosocial disease. Indeed, this suggests patients with influenza would be an excellent group to study to gain immunological insight
into the spontaneous production and remission of
mental illness.

AIDS Dementia Complex. AIDS is an infectious

disease which is characterized by a profound derangement of the immune system. As the hypothesis of this paper would predict, a global CNS
dysfunction commonly occurs with AIDS called
AIDS Dementia Complex. Usual manifestations
of AIDS Dementia Complex have been reported
to be decreased memory, inability to concentrate,
apathy, social withdrawal, psychomotor retardation, abulia, and mild headache (19). In a recent
review, Perry and Marotta wrote HIV infection
can mimic almost any neurological or psychiatric
disorder including Alzheimer disease.. . , insidious depressive withdrawal, seizures, acute florid
psychotic delirium, mania with grandiose delusions, Wernicke-Korsakoff
syndrome, with HIVinduced confusional state even before seroconversion . . . the signs and symptoms in adults described
by the numerous investigators include: forgetfulness, loss of concentration, disorientation, delirium, slowness of thought, aphasia, apraxia, deterioration of handwriting, apathy, sociaI withdrawal, dysphoric mood, vegetative depression,
delusions, paranoia, regressed behavior, organic
psychosis, somatic complaints and sociopathic behavior (20).
What causes this profound CNS derangement?
The most likely answer is increased cytokine production by immune cells, in particular, monocyte derived cells such as macrophages and microglia. This speculation is based on: 1. the powerful effects of cytokines on the CNS; 2. monocyte derived cells in the CNS of AIDS patients
are infected with HIV-l but neurons are not (21);
3. HIV-1 infected macrophages have increased
production of cytokines (22); 4. many studies
report poor correlation of behavioral pathology
with neuropathology in AIDS patients (23). This
suggests that nerve cell destruction is not the
key to AIDS Dementia Complex, which leaves
cytokines as more likely candidates, at least in
the early stages. Furthermore, the neuropathology
commonly found on autopsy of AIDS patients
could also be due, in large part, to the HIV-l infected macrophages, since they secrete increased
amounts of TNP, which has been shown to be
neurotoxic (22).

THE IMMUNE SYSTEM

IN THE ETIOLOGY

OF PSYCHOSOCIAL

One possible method of HIV-l penetraion of

the CNS is provocative. The blood-brain barrier
routinely restricts access of most molecules and
all infective agents to the brain. Blood monocytes are not restricted by the blood-brain barrier.
They pass through the blood-brain barrier into the
CNS as a normal part of immune surveillance.
In AIDS, the blood monocytes are infected with
HIV-l and when they migrate into the brain they
would most probably carry HIV-l with them. This
would seem a likely route for HIV-l infection of
the CNS.
The AIDS phenomenon suggests that blood
monocytes could be important general vehicles for
transporting infective agents and/or altered immune function from the periphery into the CNS.
If this is correct, it means that blood monocytes should be intensively studied as a possible
key factor in the pathogenesis of many human
CNS pathologies. It is possible the migration of
altered or infected monocytes from the blood
into the brain could be an essential ingredient in
the etiology of such diverse diseases as manic-depressive psychosis, schizophrenia, multiple sclerosis, Alzheimers Disease, Parkinsons Disease and
AIDS Dementia Complex.

Food and behavior

There is an extensive but neglected medical literature on food allergies as a cause of psychosocial
disease. Few if any modern textbooks on allergy
or immunology refer to this work.
Shannon (24) was the first in the American literature to describe patients with psychiatric disorders caused by food. He reported on seven high
strung, irritable, extremely nervous, restless children who also had significant sleep problems.
Elimination of various foods had remarkable behavioral benefits. Behavior deteriorated when the
offending foods were reintroduced.
Rowe was one of the most prolific and influential writers in this field. In 1930 (25) he
described a series of patients with allergic toxemia due to food allergy. This allergic toxemia
produces drowsiness, mental confusion, slowness
of thought, lack of initiative and ambition, irritability, despondency, fatigue, weakness, bodily aching and a feeling of being poisoned. He
developed a number of popular elimination diets to treat these patients. Dairy products, eggs
and wheat were the most common offenders. He

DISEASE

53

claimed his patients had dramatic benefit when the

problem foods were eliminated from their diets.
In addition to allergic toxemia he also reported
cures for various patients with migraine or epilepsy .
In 1959 (26) Rowe published his last major paper, still reporting after 30 years of clinical experience, that a wide spectrum of mental-emotional
symptoms can often be relieved by elimination diets and produced with food challenge. Fatigue was
the most common symptom and was usually accompanied by many of the following: logginess,
listlessness, drowsiness, aching muscles, irritability, tension, depression, insomnia, difficulty in
concentration and thinking.
Randolph (27) reviewed the literature on allergic toxemia and discussed five patients improved
on elimination diets from his own practice. Davison (28) coined the term cerebral allergy to descibe the phenomena he observed in 428 of his
patients. The symptoms were similar to Rowes
allergic toxemia and he reported food allergies to
be the main cause. In a later paper (29) Davison
discussed the literature on food provoked epilepsy
and reported successful outcomes using elimination diets from his own practice. Clarke (30) gave
many examples of the food-psychosocial disease
connection in a paper entitled The Relation of
Allergy to Character Problems in Children. Speer
(31) labelled the food-behavior phenomena he observed in his practice the allergic tension-fatigue
syndrome. Psychiatrists Weiss and Kaufman (32)
gave details on five young patients cured of their
emotional intensity and lability, motor hyperactiivty and confusion by eliminating certain foods.
Deamer (33), reflecting on 37 years as a pediatric
allergist, felt the allergic tension-fatigue syndrome
was one of the most underdiagnosed phenomena
in pediatrics. He reported chocolate and milk to
be the most common causes. Buisseret (34), in a
study of 79 children, found psychological disturbance was a common manifestation of cows milk
allergy.
Crook (35) reported on 50 young patients
with allergic tension-fatigue syndrome. Fatigue,
irritability, restlessness, inability to concentrate,
anxiety, tearfulness, perversity, compulsiveness,
headache and stomachache.
Elimination diets
benefited 80% of these patients. Reflecting on 20
years of pediatric practice, Crook (36) wrote he
had successfully treated over 4,000 children with
allergic tension-fatigue syndrome using elimination diets. In 1980 he titled an article, Can What

54
a Child Eats Make Him Dull, Stupid, or Hyperactive? (37). His answer was yes.
The food allergy-behavior work has been harshly
criticized (38,39,40). May called it quackery. The
claims made about allergic tension-fatigue and/or
allergic toxemia were said to be examples of imaginations deceptive powers (38). There are three
main points to the criticism: 1. misuse of the term
allergy; 2. foods claimed to cause allergic tensionfatigue are not generally supported by positive
skin tests; 3. lack of double-blinded food challenges to support the claims.
Certainly the term allergy was not used in its
strict sense, that is, an IgE mediated process, since
almost all the papers have indicated skin tests have
been useless. A better description would have been
a food-gut-immune system-behavior interaction
of unknown mechanism. The fact that skin tests
are not predictive of the food-behavior problem
should not be of concern since the food-gut-immune system-CNS axis is largely unexplored and
poorly understood. The third and most important
criticism has been answered by three landmark
controlled trials by Egger et al (41, 42, 43).
Their first paper (41) investigated 88 children
with severe migraine using an oligoantigenic elimination diet for 4 weeks followed by doubleblinded food challenge lasting one week for each
new food introduced. 93% of the children recovered on the elimination diet. The double-blinded
challenges demonstrated the effect was not due to
placebo or observer bias. In addition, 41 of the
children with migraine also had behavior disorder and 14 had epilepsy. All but 5 of the children
with behavior disorder had striking improvement
in behavior and 12 out of 14 patients with epilepsy became fit-free even after discontinuing their
medication. Double-blinded food challenges confirmed the results were not due to placebo effect
or observer bias.
Egger et als second paper (42) investigated
76 children with hyperactivity and associated
symptoms, such as fatigue, antisocial behavior,
headache and emotional difficulties. Sixty-two of
the children improved on the elimination diet and
was confirmed with double-blinded food challenge. Skin prick tests were not useful for identifying problem foods. This trial demonstrates that
diet should be taken seriously as a cause of behavior disorder in children.
Their third paper (43) compiled data on 45 epileptic children with migraine. 90% had sharp decrease in seizures during the elimination diet pe-

MEDICAL HYPOTHESES

riod. Over half ceased having seizures on the restricted diet for up to 3 years. Most had stopped
taking their medication. In the double-blinded
rechallenge, seizures were provoked in 15 out of
16 children when given a previously identified active food. No seizures were provoked with safe
foods. Children who had epilepsy without migraine showed no benefit on the elimination diet.
Egger et als work is powerful support for the
notion that food can cause psychosocial disease.
Their work, along with all the previous literature
on food-behavior, supports the hypothesis in this
paper, namely, the immune system is a key ingredient in psychosocial disease. This hypothesis
provides a much needed explanation for the findings on food-behavior, that is, food-gut interactions can activate the immune system, resulting in
the production of cytokines. The cytokines go to
receptors in the hypothalamus which profoundly
influences behavior.

Lithium and other neuroleptics

Lithium ion has revolutionized the treatment of
manic-depressive psychosis. First discovered by
Cade in 1949, it took 25 years for the US FDA
to approve it for prophylaxis against recurring
mania. Schou has suggested the long neglect was
due to its very efficiency. The effects of lithium
were thought to be too good to be true. Lithium
acts prophylactically and therapeutically against
the mania and the depressive manifestations of the
disease in over 80% of the patients. The therapeutic action is slow and less impressive than its
ability to prevent recurrence (44, 45). Lithium is
useful for sudden onset schizophrenia and schizoaffective disorders (46, 47). Furthermore, it has
been used with remarkable success on alcoholics
(48, 49). It does not cause tat-dive dyskinesia, a
common side-effect of most antipsychotic medications. Normal mood and emotional reactivity are
not affected by lithium.
Significant immune system abnormalities have
been documented in all disorders helped by
lithium. Animal studies have shown alcohol depresses immune functions, including sharp reduction in lymphocyte proliferation in response to
mitogens or recall antigens. By far the greatest
impairment occurred after alcohol administration
had ceased, corresponding to the time of alcohol
withdrawal syndrome (50). Reduced immune response is found in both well nourished and poorly
nourished alcoholics (51, 52).

THE IMMUNE SYSTEM IN THE ETIOLOGY OF PSYCHOSOCIAL

The literature on immune dysfunction in schizophrenia is difficult to evaluate since striking differences appear, depending on the medication status of the patients studied. For example, deficient IL-2 production was found in untreated
schizophrenic patients (53), but elevations were
reported in two groups of chronically medicated
schizophrenics (54, 55). Depressed levels of polymorphonuclear cells, natural killer cells, IgG, IgA
and IgM have been found along with impaired
function of natural killer cells, macrophages and
t-lymphocytes in various patient groups (56).
In mania and depression there is reduced lymphocyte beta-adrenergic receptor function (57,
58). Depressed patients appear to have reduced
mitogenic responses, but increased IgE antibodies
after exposure to antigen (59).
As early as 1973 it was discovered that lithium
activates the immune system by inducing granulopoiesis (60). It has been used to treat chemotherapy induced neutropenia (61) and there are
case studies claiming benefit for AIDS patients. A
controlled study has begun using lithium on AIDS
to combat hematocytopenias developing during zidovudine therapy (62).
Lithium is a powerful stimulator of granulopoiesis. It increases the production of colony
stimulating factors, which stimulate pluripotent
stem cells. Recent work has suggested lithium
works at a more primitive and fundamental level
than the stem cell (63). The action of lithium on
the immune system is extremely unusual, fundamental and powerful.
It is curious that after 40 years of active research on lithium, there is no understanding of
how lithium works to prevent and treat manic-depressive psychosis, alcoholism and some forms of
schizophrenia (44,45). The possibility of lithiums
beneficial action being mediated through its immune system activation has not been proposed.
It seems more than coincidental that lithium not
only activates the immune system at a fundamental level, but it also benefits three psychosocial
diseases, all three of which have documented immune dysfunction. The immune system involvement of lithium is strong support for the hypothesis of this paper. The immune system-psychosocial
disease hypothesis leads directly to the prediction
that lithiums mechanism of action is mediated
through the immune system. Lithiums long latency period before therapeutic effect is also consistent with an immune mechanism.

DISEASE

55

Other neuroleptic drugs. The slow therapeutic action of neuroleptics is one of the great paradoxes
of psychopharmacology.
Most neuroleptics block
dopamine receptors rapidly after administration,
which is thought to be their mechanism of action.
Why then does it take a week or more for the
antipsychotic effects to emerge? (64). A solution
to the paradox would be a mechanism of action involving the immune system. A surprising amount
of evidence supports that notion.
Recently, haloperidol, a calmodulin antagonist,
was shown to inhibit natural killer activity (65).
Presumably, other neuroleptics such as the phenothiazines, which are also calmodulin antagonists, would have similar effects on natural killer
activity. In an earlier experiment (66), haloperidol
and several phenothiazines blocked Epstein-Barr
virus infection of B-lymphocytes. This remarkable
immune acitivity was thought to work by blocking
calmodulin. In another study (67), the blastogenic
response of cultured T-lymphocytes was powerfully inhibited by phenothiazines, haloperidol and
thiothixene. As early as 1975 it was shown chlorpromazine inhibits concanavalin A induced lymphocyte aggregation and mitogenesis (68). In addition, benzodiazepines bind to a specific receptor
on macrophages (69).
It is evident that the major classes of neuroleptics have significant effects on the immune system.
Much more work needs to be done, but at this
point it is reasonable to propose their immune system involvement is a key, if not dominant, factor
in their antipsychotic action. If this proposal is
correct, it will also explain the long latency of action paradox and eventually lead to a comprehensive understanding of the etiology and treatment
of psychosocial disease.
Conclusion
The astonishing effects of interleukin-2 and interferon on the CNS provide a solid foundation
for the assertion that the immune system plays
a key role in the etiology of psychosocial disease. Consistent with and supportive of that hypothesis is a remarkable amount of diverse data
on: food allergy; lithium and other neuroleptics;
symptoms of infectious disease; AIDS Dementia Complex; blood monocyte migration into the
CNS; the dense two-way communications network
between the CNS and the immune system; and the
direct access of cytokines to the hypothalamus,

56
the premier control center of the brain. This hypothesis provides much needed explanations for:
1. the powerful effect of food on behavior; 2.
the mechanism of action of lithium carbonate and
other antipsychotic drugs; 3. the delayed therapeutic action of neuroleptics; and 4. the rapid onset of AIDS Dementia Complex. This hypothesis,
if correct, will usher in a golden age of psychosocial disease research. The full power of modern
biomedical science could be put into action, since
the immune system is largely blood borne. No
longer will the inaccessability of the brain be the
great barrier to research on psychosocial disease.
References
1. Mitchinson MJ, Ball RY. Macrophages and atherosclerosis. Lancet 2: 146, 1987.
2. Balkwill FR, Burke F. The cytokine Network. Immunol
Today 10: 299. 1989.
3. Hucklebridge FH et al, Methionine-enkephalin stimulates
in vitro proliferation of human peripheral lymphocytes via
&opioid receptors. Brain Behavior Immunity 3: 183.1989.
4. Morgan0 A et al. Expression of HLA-class II antigens and
proliferative capacity in autologous mixed lymphocyte reactions of human T lymphocytes exposed in vitro to a-Endorphin. Brain Behavior Immunity 3: 214, 1989.
5. Maslinski W. Cholinergic receptors of Lymphocytes.
Brain Behavior Immunity 3: 1, 1989.
6. Goetzl E, Sreedharan SP, Harkonen WS. Pathogenic
roles of neuroimmunologic mediators. Immunol Allergy
Clin N Amer 8: 183, 1988.
7. Besedovsky H, de1 Rey A, Sorkin E. Immunological-neuroendocrine feedback circuits, p 165 in Neural Modulation of Immunity (R. Guillemin et al, eds) Raven Press,
New York, 1985.
8. Hall NR, Goldstein AL. Nemotransmitters and host defense, p 143 in Neural Modulation of Immunity (R
Guillemin et al, eds) Raven Press, New York, 1985.
9. Plata-Salaman CR. Immunomodulators and feeding regulation: a humoral link between the immune and nervous
systems. Brain Behavior Immunity 3: 193, 1989.
10. Shepherd GM. Neurobiology, 2nd ed. Oxford University
Press, New York, 1988.
11. Thompson RF. The Brain. WH Freeman and Company,
New York, 1985.
12. ibid, p 19.
13. Denicoff KD. The neuropsychiatric effects of treatment
with interleukin-2 and lymphokine-activated killer cells.
Ann Int Med 107: 293, 1987.
14. ibid, p 299.
15. Fent K, Zbinden G. Toxicity of interferon and interleukin.
Trends Pharmacol Sci 8: 100, 1987.
16. Rohatiner AZS et al. Central nervous system toxicity of
interferon. Br J Cancer 47: 419, 1983.
17. Adams F, Quesada JR, Gutter-man JU. Neuropsychiatric
manifestationos of human leukocyte interferon therapy in
patients with cancer. J Am Med Assn 252: 938, 1984.
18. Zisook S. DeJirium. Psychiatric Med 6: 8. 1988.
19. Bredesen DE. Clinical features, p 401, In: Ho DD, moderator. The acquired immunodeficiency syndrome (AIDS)
dementia complex. Ann Intern Med 111: 400, 1989.

MEDICAL HYPOTHESES

20. Perry S, Marotta RF. AIDS dementia: a review of the literature. Alzheimer Dis Assoc Disorders 1: 221, 1987.
21. Ho DD, Daar ES. Neurovirology and neuropathogenesis,
p 405, In Ho DD, moderation. The acquired immunodeficiency syndrome (AIDS) dementia complex. Ann Intern
Med 111: 400, 1989.
22. ibid p 407.
23. Navia BA, Cho ES, Petit0 CK, Price RW. The AIDS
dementia complex: II. Neuropathology. Ann Neurol 19:
525, 1986.
24. Shannon WR. Neuropathic manifestations in infants and
children as a result of anaphylactic reaction to foods contained in their dietary. Am J Dis Child 24: 89, 1922.
25. Rowe AH. Allergic toxemia and migraine due to food allergy. Cal West Med 33: 785, 1930.
26. Rowe AH. Allergic fatigue and toxemia. Ann Allergy 17:
9, 1959.
27. Randolph TG. Fatigue and weakness of allergic origin (allergic toxemia) to be differentiated from nervous fatigue
or neurasthenia. Arm Allergy 3: 418, 1945.
28. Davison HM. Cerebral allergy. South Med J 42: 712,
1949.
29. Davison HM. Allergy of the nervous system. Quart Rev
Allergy Appl Immuno16: 157, 1952.
30. Clarke TW. The relation of allergy to character problems
in children: a survey. Psychiatric Quart 24: 21, 1950.
31. Speer F. Allergic tension-fatigue in children. Ann Allergy
12: 168, 1953.
32. Weiss JM, Kaufman HS. A subtle organic component in
some cases of mental illness. Arch Gen Psychiat 25: 74,
1971.
33. Deamer WC. Pediatric allergy: some impressions gained
over a 37 year oeriod. Pediatrics 48: 930. 1971.
34. Buisseret PD. Common manifestations of cows milk allergy in children. Lancet 1: 304, 1978.
35. Crook WG, Harrison WW, Crawford SE, Emerson BS.
Systemic manifestations due to allergy. Pediatrics 27: 790,
1961.
36. Crook WG. Food allergy-the great masquerader. Pediatric Clin N Amer 22: 227, 1975.
37. Crook WG. Can what a child eats make him dull, stupid,
or hyperactive? J Learn Disabilities 13: 53, 1980.
38. May CD. Food allergy. Pediatric Clin N Amer 22: 217,
1975.
39. May CD. Food allergy-material and ethereal. New Engl J
Med 39: 1142, 1980.
40. May CD. food allergy: lessons from the past. J Allergy
Clin Immuno169: 255, 1982.
41. Egger J et al. Is migraine food allergy? Lancet 2: 865,
1983.
42. Egger J et al. Controlled trial of oligoantigenic treatment
in the hyperkinetic syndrome. Lancet 1: 540, 1985.
43. Egger J, Carter CM, Soothill JF, Wilson J. Oligoantigenic
diet treatment of children with epilepsy and migraine. J
Pediatrics 114: 51, 1989.
44. Schou M. Lithium treatment of manic-depressive illness.
J Am Med Assn 259: 1834, 1988.
45. Stern TA, Lydiard RB. Lithium therapy revisitetd. Psychiatric Med 4: 39, 1987.
46. Delva NJ, Letemendia FJJ. Lithium treatment in schizophrenia schizo-affective disorders. Brit J Psychiat 141:
387, 1982.
41. Hirschowitz J, Casper R, Garver DL, Chang S. Lithium
response in good prognosis schizophrenia. Am J Psychiatry 137: 916, 1980.

THE IMMUNE SYSTEM IN THE ETIOLOGY OF PSYCHOSOCIAL

48. Fawcett J et al. A double-blind, placebo-controlled trial

of lithium carbonate therapy for alcoholism. Arch Gen
Psychiatry 44: 248, 1987.
49. Flemenbaum A. Lithium carbonate prophylaxis of alcoholism: its time has come. Arch Gen Psychiatry 46: 290,
1989.
50. Jerrells TR et al. Ethanol-associated immunosuppression.
p 173 in Psychological, Neuropsychiatric and Substance
Abuse Aspects of AIDS. (Bridge TP et al, eds) Raven
press, New York, 1988.
51. Stefanini GF et al. In vivo effect of chronic ethanol abuse
membrane alpha 1-glycoprotein of lymphocytes and immune response to various stimulating agents. Alcoholism:
Clin Exp Res 13: 444, 1989.
52. Morland H et al. Reduced IgG Fc-receptor-mediated
phagocytosis in human monocytes isolated from alcoholics. Alcoholism: Clin Exp Res 12: 755, 1988.
53. Villemain F et al. Aberrant T cell-mediated immunity in
untreated schizophrenic patients: deficient interleukin-2
production. Am J Psychiatry 146: 609, 1989.
54. Rapaport MH et al. Elevated levels of soluble interleukin
2 receptors in schizophrenia. Arch Gen Psychiatry 46:
291, 1989.
55. Ganguli R, Rabin BS. Increased serum interleukin 2 receptor concentration in schizophrenic and brain-damaged
subjects. Arch Gen Psychiatry 46: 292, 1989.
56. DeLisi LE. Is immune dysfunction associated with
schizophrenia. Psychopharacol Bull 20: 509, 1984.
51. Extem I, Talbnan J, Smith CC, Goodwin FK. Changes
in lymphocyte beta-adrenergic receptors in depression and
mania. Psychiatry Res 1: 191, 1979.
58. Mann JJ et al. Reduced sensitivity of lymphocyte betaadrenergic receptors in patients with endogenous depression and psychomotor agitation. N Engl J Med 313: 715,
1985.
59. Lieb J. Lithium and immune function. Med Hypoth 23:
73. 1987.

DISEASE

51

60. Tisman G, Herbert V, Rosenblatt S. Evidence that lithium

61.

62.

63.

64.

65.

66.

61.

68.

69.

induces human granulocyte proliferation: elevated serum

B12 binding capacity in vivo and granulocyte colony proliferation in vitro. Br J Haematol 24: 767, 1973.
Richman CM, Makii MM, Weiser PA, Herbst AL. The
effect of lithium carbonate on chemotherapy-induced neutropenia and thrombocytopenia. Am J Hematol 16: 313:
1984.
Herbert V, Hirschman S, Jacobson J. Lithium for
zidovudine-induced neutropenia in AIDS. J Am Med
Assn 260: 3588, 1988.
Doukas MA, Shadduck RK, Waheed A, Gass C. Lithium
stimulation of diffusion chamber colony growth is mediated by factors other than colony-stimulating factor. Int
J Cell Clon 7: 168, 1989.
Freed WJ. The therapeutic latency of neuroleptic drugs
and nonspecific postjunctional supersensitivity. Schizophrenia Bull 14: 269, 1988.
Solovera JJ et al. Inhibition of B lymphocytes by a human
herpesvirus, Epstein-Barr virus, is blocked by calmodulin
antagonists. Proc Nat1 Acad Sci USA 81: 4955, 1984.
Nemerow GR, Cooper NR. Infection of B lymphocytes
by a human herpesvirus, Epstein-Barr virus, is blocked
by calmodulin antagonists. Proc Nat1 Acad Sci USA 81:
4955, 1984.
Baker GA, Santalo R, Blumenstein J. Effect of psychotropic agents upon the blastogenic response fo human
T-lymphocytes. Biol Psychiatry 12: 159, 1977.
Ferguson RM, Schmidtke JR, Simmons RL. Concurrent inhibition by chlorpromazine of concanavalin A-induced lymphocyte aggregation and mitogenesis. Nature
256: 744, 1975.
Zavala F, Lenfant M. Benzodiazepines and PK 11195exert immunomodulating activities by binding on a specific
receptor on macrophages. Ann NY Acad Sci 4%: 240,
1987.