Psychoneuroimmunology Research Paper

First, however, a brief reminder that the immune system’s defense of the organism against foreign, ‘nonself’ material (antigens) is executed by white blood cells, primarily T and B lymphocytes, capable of responding clonally to antigens and retaining the ‘memory’ of that encounter. Antibody-mediated immunity refers to the exposure of antigens to bonemarrow derived B cells which produce antibodies that protect the organism against extracellular infectious agents. Cell-mediated immunity is provided by thymus-derived T cells that defend against intracellular parasitic and viral infections. An integrated immune response actually involves interactions among specialized subpopulations of T cells (e.g., helper, cytotoxic), B cells, other white blood cells such as macrophages, and substances (cytokines) secreted by activated immune cells. Lymphocyte proliferation in response to in itro stimulation by mitogens (plant lectins) that preferentially influence T and/or B cells is a measure of the capacity of these lymphocytes to divide in response to nonspecific foreign stimuli. Natural killer (NK) cells are large, granular lymphocytes that are implicated in tumor metastases and in the recognition and defense against viral agents without having had prior experience with that antigen.

1. Historical Perspective

The first sustained program of research on brainimmune system interactions was initiated by Russian investigators in the 1920s. Derived directly from a Pavlovian perspective, these investigators studied the conditioning of immune responses. There was reliable evidence for conditioned alterations in nonspecific immunologic reactions; the evidence for a conditioned change in antibody production was less convincing (Ader 1981b). Early studies on the effects of brain lesions on anaphylactic responses began in the 1950s and were pursued by investigators in the former Soviet Union and in the United States who studied the effects of hypothalamic lesions and electrical stimulation on several parameters of immunologic reactivity.

In the 1950s and 1960s, Fred Rasmussen, Jr., a virologist, and Norman Brill, a psychiatrist (the first such interdisciplinary collaboration), began a series of studies on the effects of stress (avoidance conditioning, physical restraint, electric shock, social ‘crowding’) on susceptibility to experimentally-induced infectious diseases (e.g., Rasmussen et al. 1957). Depending on the nature of the stressor, these stimuli increased susceptibility to some viral infections and decreased susceptibility to others. These experiments stimulated only a few other investigators to study the effects of early life experiences and stressful psychosocial circumstances on susceptibility to infectious and neoplastic diseases in animals.

In the 1970s, things began to change. Several dramatic observations led to sustained research on the relationship between the brain and the immune system. Hadden et al. (1970) documented the existence of receptors on lymphocytes, providing the first observations linking the immune system to the sympathetic nervous system; Ader and Cohen (1975) demonstrated Behaviorally conditioned suppression of the immune system; Bartrop et al. (1977) described immunologic effects associated with the bereavement that followed the death of a spouse; Besedovsky and Sorkin (1977) began their studies of a neuroendocrine-immune system network and the effects of immune responses on neural and endocrine function; the sympathetic innervation of lymphoid tissues was documented (Felten and Felten 1991); and Blalock and Smith (1980) showed that lymphocytes, themselves, were capable of producing neuropeptides. These seminal studies provided the impetus for new research on receptors for hormones and neurotransmitters on lymphocytes, the effects of sympathetic nervous system (SNS) activity on immunity, conditioned immune responses, depression and immunity, the effects of stressful life experiences on immune function and resistance to disease, psychosocial factors in the progression of AIDS, and immunologic effects on Behavior.

2. Brain–Immune System Communication

Two pathways link the brain and the immune system: autonomic nervous system activity and neuroendocrine outflow via the pituitary. Primary (bone marrow, thymus) and secondary (e.g., spleen, lymph nodes) lymphoid organs are innervated with sympathetic nerve fibers that form close neuroeffector junctions with lymphocytes and macrophages and release a variety of neuropeptides. The stimulation or interruption of these connections can influence immune responses. Chemical sympathectomy, for example, suppresses antibody and cell-mediated immunity; hypothalamic lesions generally inhibit and stimulation potentiates immunologic reactivity.

Conversely, activation of the immune system is accompanied by changes in hypothalamic, autonomic, and endocrine processes. There is, for example, an increase in the firing rate of neurons in the hypothalamus that corresponds to the time of peak antibody production; sympathetic activity (turnover of norepinephrine) is increased in the spleen and hypothalamus; and immune responses initiated by viral infections are associated with increases in circulating levels of ACTH and corticosterone.

Because lymphocytes carry receptors for a variety of hormones and neuropeptides, the cellular interactions that mediate humoral and cellular immunity can be modulated by the neuroendocrine environment in which these responses occur. Lymphocytes bear receptors for corticotropin-releasing factor (CRF), ACTH, and endogenous opioids. There are direct immunomodulatory effects of CRF and ACTH, but their major in i o effects are exerted through interactions with other hormones and immune system products, the most conspicuous of which is the release of adrenocortical steroids. There is an extensive literature documenting the anti-inflammatory and immunosuppressive effects of corticosteroids—at least in pharmacologic doses. At physiologic doses, adrenocortical steroids are essential for normal immune function and, depending on steroid concentration, antigen concentration, and the temporal relationship between hormonal and antigenic stimulation, corticosteroids can be immunoenhancing.

Recent data have also shown that lymphocytes activated by antigenic stimuli are capable of producing neuropeptides and hormones. Thus, the brain is capable of detecting and responding to signals released by an activated immune system. Cytokines, nonantibody messenger molecules released by activated immune cells, besides regulating cellular interactions, are another means by which the immune system communicates with the CNS and, thereby, influences Behavior. The precise site(s) at which cytokines act within the brain has not been elaborated. Nonetheless, it is known that cytokines are electrophysiologically, endocrinologically and Behaviorally active. Symptoms of sickness, for example, decreased activity, loss of interest in the environment, reduced food intake, fatigue, and cognitive disorders and affective changes can be induced in healthy subjects by the central or peripheral injection of proinflammatory cytokines. Indeed, the recent therapeutic use of interferons in human disease has been associated with neurologic and psychiatric side effects (Meyers 1999).

At the neural and endocrine levels, then, there is abundant evidence of interactions between the brain and the immune system.

3. Behavioral Influences On Immune Function

3.1 Effects Of ‘Stress’ On Immune Function

The link between Behavior and immune function goes back to the earliest observations of a relationship between psychosocial factors and susceptibility to immunologically mediated disease processes. There is compelling evidence that psychosocial factors influence the onset or severity of rheumatoid arthritis and other autoimmune diseases, diseases effected by immune reactions generated against an individual’s own tissues. Similarly, stressful life experiences are associated with increased susceptibility to infectious diseases and the reactivation of latent viral infections. The death of a spouse, for example, has been associated with depression and an increased morbidity and mortality in response to diseases presumed to involve immune defenses. Bereavement and/or depression is also accompanied by a depression of immune function (reduced lymphoproliferative responses, impaired NK cell activity). In none of these instances, however, has it been demonstrated that these health effects are the result of stressor-induced changes in immune function.

Other ‘losses,’ such as marital separation and divorce and other chronic stressors such as caregiving for disabled patients, have also been associated with decreases in lymphocyte function and NK cell activity and a depressed antiviral response to an influenza immunization. In individuals seropositive for EpsteinBarr virus (EBV), there is also an increase in anti-EBV antibody titers reflecting compromised cell-mediated immune system control over the latent virus. Even more seemingly innocuous circumstances such as taking examinations results in a transient increase in levels of distress and a corresponding decrease in lymph proliferative responsiveness, NK cell activity, percentage of helper T-lymphocytes, and interferon production by stimulated lymphocytes. Exam periods are also associated with elevated EBV titers and an increase in the incidence of self-reported symptoms of infectious illness.

Clinical studies have described an association between some immunologically mediated diseases and clinical depression. Multiple sclerosis (MS), a T cell mediated autoimmune disease, is associated with depressive states. Systemic lupus erythematosus (SLE) is a B cell mediated autoimmune disease in which depression is frequently reported. Depressive features have also been described in HIV-infected individuals. Thus, considerable attention has been given to the effects of clinical depression on immune function. Based on a meta-analysis of the literature (Herbert and Cohen 1993), depression is reliably associated with a decline in both enumerative and functional measures of immunity such as lymph proliferative responses and NK cell activity. Depressed patients also show elevated antibody titers to herpes simplex virus (HSV) and cytomegalovirus and a decreased ability to mount a specific cell-mediated response to varicella zoster virus. The decline in cellular immunity is comparable to the decline observed in elderly patients who are more susceptible to herpes zoster infection than younger adults (Irwin et al. 1998). Other measures of immune function in depressed patients are characterized by a high variability within and between studies.

Typically, acute laboratory stressors such as unsolvable puzzles or public speaking have been reported to increase numbers of cytotoxic lymphocytes and the number and activity of NK cells along with a decrease in lymph proliferative responses. These data, too, are characterized by high variability that is, at least in part, related to individual differences in autonomic nervous system reactivity. For example, subjects who are characterized as high SNS reactors in response to an acute laboratory stressor show a greater decrease in lymph proliferative responses, a greater increase in the number of suppressor cytotoxic T cells and, perhaps, a greater increase in NK cell activity than low reactors.

It is clear that, while stressful life experiences influence immune function in humans, there are large individual differences due, presumably, to personality and dispositional factors that influence the individual’s perception and appraisal of the environmental circumstances, the individual and social resources available to cope effectively with the demands of the environment, and the biological substrate upon which the stressful stimuli are superimposed.

Most of the evidence for stress-induced alterations in immunity comes from basic research on animals. Stressors such as restraint, electric shock, avoidance conditioning, and noise can disrupt immune function and increase susceptibility and/or mortality to a variety of infectious agents (e.g., HSV, influenza, Coxsackie virus, Sindbis virus) or allow an otherwise inconsequential exposure to a pathogen to develop into clinical disease (Moynihan and Ader 1996, Sheridan et al. 1994).

In contrast, though, some of these same stressors decreased susceptibility to poliomyelitis virus infection, experimental allergic encephalomyocarditis and Plasmodium berghei, a rodent malaria. Stressful life experiences also exert a protective effect on the development of experimental allergic encephalomyelitis, a murine model of MS. In the case of experimentally induced arthritis, increased or decreased susceptibility depends on the experimental model of arthritis and the nature of the stressor.

Prenatal and early life experiences such as interruptions of mother-young interactions, the social environment in which animals live, exposure to predators or pheromones emitted by stressed conspecifics, and subjecting animals to environmental conditions over which they have no control (e.g., physical restraint, electric foot shocks, forced exercise) are among the psychological and social manipulations that alter neuroendocrine states and modulate immune responses. Antibody production is generally suppressed by such stressful stimulation.

Although differential housing (frequently mislabeled as ‘crowding’ or ‘isolation’) may not qualify as a stressor, the manner in which animals are housed or a change in housing conditions is sufficient to influence the antibody response to different antigens. Primary and/or secondary antibody responses in rodents are suppressed by restraint or electric shock stimulation, depending on the nature of the antigen and the timing of the stressor in relation to the antigenic stimulation. Other stressors, however, are associated with enhanced antibody production. Similarly, cell mediated processes (e.g., graft rejection, delayed type hypersensitivity reactions) and nonimmunologically specific responses (e.g., lymphocyte proliferation in response to mitogens, NK cell activity, cytokine production) generally are suppressed in animals subjected to stressors but, again, the nature and timing of the stressor, the particular response being measured as well as the species, strain and sex of the animal will determine the direction and magnitude of the effects.

A few attempts have been made to discriminate between stressors over which the individual does or does not have control, for example, avoidable vs. unavoidable, escapable vs. inescapable, signaled vs. unsignaled electric shock in animals. Although differences have been observed in some studies, the results have been contradictory. In humans, too, studies involving control—or perceived control—over stressful stimulation have yielded inconsistent results. Given the number of variables identified as having an effect on the assessment of immunologic reactivity, however, there are insufficient data to warrant any conclusions at this time.

While many hypotheses have been proffered, the neuroendocrine mechanisms presumed to mediate the effects of stressful life experiences on immune function remain elusive. For one thing, our understanding of neuroendocrine-immune system interactions under normal and stressful conditions is incomplete. Glucocorticoids, for example, are generally immunosuppressive, and it is generally assumed that corticosteroid elevations, the most common manifestation of the stress response are responsible for the immunosuppression associated with stressful life experiences. There are many examples of stress-induced, corticosteroid-mediated alterations of immune responses. But, there are very many other observations of stressinduced alterations of immunity that are independent of adrenocortical activity. Intermittent and continuous schedules of inescapable electric foots-hocks, both of which elevate corticosterone levels, effect an analgesic response to subsequent foot-shock. Only the intermittent shock, however, is an opioid-mediated analgesia and only the intermittent foot-shock suppresses NK cell activity, illustrating the immunomodulating potential of endogenous opioids.

Even without being able to detail the extensive literature, it is clear that the response to stressful life experiences involves complex neural, endocrine, and immune response interactions (Glaser and KiecoltGlaser 1994, Rabin 1999). Different stimuli commonly referred to as stressors—and commonly thought to elicit a representative ‘stress’ response—can have different effects on the same immune response. Conversely, the same stressor can have different effects on different immune responses. Also, the intensity and chronicity of the stressor are among the several parameters of stimulation which influence immune responses. In general, then, we have an incomplete picture of the direction, magnitude, and duration of the effects of stressful life experiences on immunity. We do, however, know something about the factors that contribute to the observed variability. They include:

(a) the quality and quantity of stressful stimulation;

(b) the capacity of the organism to cope effectively with stressful circumstances;

(c) the quality and quantity of immunogenic (or pathogenic) stimulation;

(d) the temporal relationship between stressful stimulation and immunogenic stimulation;

(e) the parameters of immune function and the sample times chosen for measurement;

(f) the experiential history of the organism and the prevailing social and environmental conditions upon which stressful stimulation and immunogenic stimulation are superimposed;

(g) a variety of host factors such as species, strain, age, sex, nutritional state and current disease; and

(h) interactions among these several variables.

3.2 Conditioning Effects

Behaviorally conditioned alterations of immune function provide one of the more dramatic lines of evidence linking the brain and the immune system. In a prototypical study (Ader and Cohen 1975), a novel, distinctively-flavored drinking solution (e.g., saccharin), the conditioned stimulus (CS) is paired with the injection of an immunomodulating drug, the unconditioned stimulus (UCS). When animals are subsequently injected with antigen, some of them are re-exposed to the CS, alone. When the immunosuppressive drug, cyclophosphamide (CY), was used as the UCS, conditioned animals re-exposed to the CS showed an aversion to the saccharin solution and an attenuated antibody response relative to conditioned animals that were not re-exposed to the CS or nonconditioned animals that were exposed to saccharin. Several independent investigators have verified and extended these findings.

The effects of conditioning have been consistent under a variety of experimental conditions (Ader and Cohen 2001). Studies have documented the acquisition and/or extinction of conditioned nonspecific host defense responses such as NK cell activity and different cell-mediated as well as antibody-mediated immune responses using different unconditioned and conditioned stimuli. Also, conditioning is not limited to changes associated with taste aversion learning; several studies have described the acquisition and extinction of conditioned ‘stress’ effects. Moreover, there is no consistent evidence of a relationship between conditioned Behavioral and conditioned immune changes. Taste aversions can be expressed without concomitant changes in immune function, and conditioned changes in immunologic reactivity can be obtained without observable avoidance conditioning. Also, the available data indicate that conditioned immunosuppressive responses cannot be attributed to either stress-induced or conditioned elevations of adrenocortical steroids.

The acquisition and extinction of a conditioned enhancement of immunologic reactivity have been observed using antigens rather than pharmacologic agents as UCSs. In the first such study (Gorczynski et al. 1982), CBA mice were anesthetized, shaved, grafted with skin from C57BL/6J mice, and left bandaged for 9 days. In response to the grafting of allogeneic tissue, there was, as expected, an increase in the number of precursors of cytotoxic T lymphocytes (CTLp) that could react against the foreign (nonself ) tissue. The conditioning procedures were repeated when CTLp numbers returned to baseline levels. On the fourth such trial, the procedures were repeated again, but the tissue graft was omitted. Approximately half the animals in each of several experiments showed an increase in CTLp in response to the grafting procedures alone. Half of the ‘responder’ mice were given additional conditioning trials and half received extinction trials (sham grafting). When tested again, all the animals given additional conditioning trials showed the conditioned increase in CTLp whereas none of the ‘responder’ mice that experienced unreinforced trials showed a conditioned response.

In another study on conditioned immunoenhancement, mice were immunized repeatedly with low doses of a common laboratory antigen following exposure to a gustatory CS. Three weeks after the last trial, half the animals in each of the several groups were re-exposed to the CS, alone, or were re-exposed to the CS in the context of a low-dose booster injection of the same antigen. The latter groups were added to determine if conditioning could be more sensitively indexed by asking if a CS could potentiate the effects of a minimally effective dose of antigen rather than initiate the production of antibody by itself. In this experiment, neither the CS nor the booster dose of antigen, alone, were sufficient to elicit a robust antibody response. However, antibody production was enhanced when conditioned mice were re-exposed to the CS in the context of re-exposure to a minimally immunogenic dose of that same antigen. Recently, others have observed a conditioned enhancement of antibody production after a single conditioning trial and without the need for a booster injection of antigen.

Data on conditioning in humans is limited. The anticipatory (conditioned) nausea that frequently precedes immunosuppressive cancer chemotherapy was also associated with anticipatory suppression of proliferative responses to T cell mitogens. In healthy subjects, the pairing of a distinctive flavor with injections of adrenaline resulted in an enhancement of NK cell activity when subjects were subsequently re-exposed to the CS alone. Preliminary data on cell mediated immunity derived from a study in which subjects were repeatedly injected with tuberculin obtained from a green vial and with saline drawn from a red vial. On the test trial, the contents of the vials were switched. Saline administered to the arm previously treated with tuberculin did not elicit a skin reaction, but there was an attenuated response in the arm previously injected with saline. These studies lend credence to earlier anecdotal reports of what would now appear to be examples of conditioned alterations of immunologic reactivity in human subjects or patients.

4. Biologic Impact Of Behaviorally-Induced Alterations Of Immune Function

Many studies have related stressful life experiences to changes in immunity and/or changes in the onset or progression of disease (Biondi and Zinnino 1997), and these correlational studies are, indeed, provocative. The association between stressful life events and increased morbidity and the association between stressful life events and alterations in immune function do not, however, establish a causal chain linking stressful life events, immune function, and health or disease. There are, at present, few, if any, human studies in which altered resistance to disease has been demonstrated to be a direct result of biologically relevant changes in immune function induced by psychosocial or stressful life experiences. Such data are being collected from animal models of spontaneously occurring disease or models in which it is possible to precipitate adaptive responses to immunogenic stimuli experimentally.

In mice infected with influenza virus, for example, there is a delay in the production of virus-specific antibody in restrained relative to unrestrained animals. In mice inoculated with HSV, prolonged periods of restraint or repeated periods of electric shock suppressed NK cell activity and the development of HSV-specific cytotoxic T lymphocytes. Also, higher titers of infectious virus were recovered from the restrained mice. Other recent studies have shown that psychosocial factors can reactivate latent HSV. While restraint was ineffective, disruption of the social hierarchy within a colony of mice by repeatedly introducing an intruder into a group cage increased aggressive Behavior, activated the hypothalamic–pituitary–adrenal axis (HPA), and resulted in reactivation of latent HSV in more than 40 percent of the latently infected animals.

In an NK-sensitive tumor model in which lung metastasis is related to NK cell function, it has been found that, depending upon when the stressor was imposed in relation to the effect of NK cells on the metastatic process, forced swimming, acute alcohol intoxication and social confrontations decreased NK cell activity and increased lung metastases. This finding is consistent with strain, age, and sex differences in NK cell activity and the development of lung metastases.

In defending the organism against infection, inflammatory processes are an essential component in the healing of wounds. Both the SNS and the HPA axis are capable of modulating inflammatory and immune processes. Thus, stressful life experiences would be expected to influence wound healing as, in fact, it does. Chronically stressed caretakers of Alzheimer’s patients took 24 percent longer than control subjects to heal a standard, experimentally produced wound in the skin, and there was a 40 percent delay in wound healing in dentistry students tested just before an examination period (when perceived stress was high) compared to healing during a summer vacation (when perceived stress was low). Paralleling this human research, recent animal studies have found that mice restrained for 3 days before and 5 days after wounding showed a reduced inflammatory response, an elevated corticosterone level and a dramatic delay in the rate at which the wounds healed.

These animal models of wound healing, NK-sensitive tumors and reactivation of latent virus infection will enable research designed to examine the mechanisms underlying the effects of stressful life experiences and determine whether they are, as suggested, due to Behaviorally-induced alterations in immune function.

The spontaneous development of a systemic lupus erythematosus-like disease in New Zealand mice was used to assess the biologic impact of conditioned alterations in immune responses since a suppression of immunologic reactivity would be in the biological interests of these animals. Half the weekly treatments received by the lupus-prone mice consisted of saccharin plus an ip injection of saline (the CS) rather than the CS plus an injection of cyclophosphamide. Under this pharmacotherapeutic regimen, the onset of disease was delayed using a cumulative amount of immunosuppressive drug that was not, by itself, sufficient to alter progression of the autoimmune disorder.

Consensual validity for these observations was provided by a study in which re-exposure to a CS previously associated with CY accelerated mortality in conditioned mice in response to a transplanted plasmacytoma. It has also been shown that protection against adjuvant-induced arthritis could be achieved by exposing animals to a CS previously paired with an immunosuppressive drug or with a stressful (immunosuppressive) experience. Among mice previously conditioned by the pairing of a CS with an immunosuppressive drug, re-exposure to the CS following transplantation of allogeneic tissues prolonged survival of the allografts. Clearly, conditioned immunologic effects can have a profound biologic impact on the development of autoimmune disease, neoplastic processes, and graft survival. Thus far, there is only one clinical case study in which a conditioning protocol was successfully used to reduce the total amount of cytoxan used in the treatment of a child with SLE (Olness and Ader 1992).

5. Conclusions

‘Psychoneuroimmunology’ is an interdisciplinary field of study that has developed and now prospers by exploring and tilling fertile territories secreted by the arbitrary if not illusory boundaries of the biomedical sciences. Disciplinary boundaries and the bureaucracies they created are biological fictions that can restrict imaginative research and the transfer and application of technologies. They lend credence to Heisenberg’s assertion (1958) that ‘What we observe is not nature itself, but nature exposed to our method of questioning’ (Ader 1995).

A paradigm shift is occurring, however. Based on research conducted since the mid-1970s, it has become evident that immunoregulatory processes can no longer be studied as the independent activity of an autonomous immune system. We now know of at least two pathways through which the brain communicates with the immune system: autonomic nervous system activity and neuroendocrine outflow via the pituitary. Both pathways generate chemical signals that can be perceived by the immune system via receptors on the surface of lymphocytes and other immune cells. Conversely, an activated immune system generates chemical signals that can be perceived by the nervous system. Thus, there is a bidirectional exchange of information between the brain and the immune system and all immunoregulatory processes take place within a neuroendocrine milieu that is responsive to the individual’s perception of and adaptive responses to physical and psychosocial events occurring in the external world.

It should not be surprising, then, that stressful life experiences or Pavlovian conditioning could influence immune function. The conditioned suppression or enhancement of immune responses has been demonstrated in different species and under a variety of experimental conditions. Also, stressful life experiences influence immune function. Both conditioning and stressful stimulation exert biologically meaningful effects in that they appear to be implicated in altering the development and/or progression of immunologically mediated disease processes. While these are reproducible phenomenon, the extent to which one can generalize from one stressor or from one parameter of immune function to another is limited. Also, the direction and/or magnitude of the effects of Behavioral factors in modulating immune responses depend on the nature of the Behavioral circumstances; the nature of the antigenic stimulation and the temporal relationship between them; the immune response and time at which it is measured; a variety of host factors; and the interactions among these variables.

The neuroendocrine pathways underlying the effects of conditioning or stressful life experiences on immune responses are not yet known. It would appear, though, that the neural and endocrine changes associated with changes in Behavioral states and the network of connections between the brain and the immune system provide multiple pathways through which Behavioral processes could influence immune defenses. The existence of these bidirectional pathways reinforces the hypothesis that immune changes constitute an important mechanism through which psychosocial factors could influence health and disease.

Bibliography:

1. Ader R (ed.) 1981a Psychoneuroimmunology. Academic Press, New York
2. Ader R 1981b A historical account of conditioned immunobiologic responses. In: Ader R (ed.) Psychoneuroimmunology. Academic Press, New York
3. Ader R 1995 Historical perspectives on psychoneuroimmunology. In: Friedman H, Klein T W, Friedman A L (eds.) Psychoneuroimmunology, Stress and Infection. CRC Press, Boca Raton, FL
4. Ader R, Cohen N 1975 Behaviorally conditioned immunosuppression. Psychosomatic Medicine 37: 333–40
5. Ader R, Cohen N 2001 Conditioning and immunity. In: Ader R, Felten D L, Cohen N (eds.) Psychoneuroimmunology, 3rd edn., Academic Press, New York
6. Bartrop R W, Lazarus L, Luckhurst E, Kiloh L G, Penny R 1977 Depressed lymphocyte function after bereavement. Lancet 1 i: 834–37
7. Besedovsky H, Sorkin E 1977 Network of immune-neuroendocrine interactions. Clinical and Experimental Immunology 27: 1–12
8. Biondi M, Zannino L-G 1997 Psychological stress, neuroimmunomodulation, and susceptibility to infectious diseases in animals and man: A review. Psychotherapy and Psychosomatics 66: 3–26
9. Blalock J E, Smith E M 1980 Human leukocyte interferon: Structural and biological relatedness to adrenocorticotropic hormone and endorphins. Proceedings of the National Academy of Science, USA 77: 5972–74
10. Felten S Y, Felten D L 1991 Innervation of lymphoid tissue. In: Ader R, Felten D L, Cohen N (eds.) Psychoneuroimmunology, 2nd edn. Academic Press, New York
11. Glaser R, Kiecolt-Glaser J K (eds.) 1994 Handbook of Human Stress and Immunity. Academic Press, San Diego, CA
12. Gorczynski R M, Macrae S, Kennedy M 1982 Conditioned immune response associated with allogeneic skin grafts in mice. Journal of Immunology 129: 704–9
13. Hadden J W, Hadden E M, Middleton E 1970 Lymphocyte blast transformation. I. Demonstration of adrenergic receptors in human peripheral lymphocytes. Journal of Cellular Immunology 1: 583–95
14. Heisenberg W 1958 Physics and Philosophy: The Revolution in Modern Science, 1st edn. Harper, New York
15. Herbert T B, Cohen S 1993 Depression and immunity: A meta-analytic review. Psychology Bulletin 113: 472–86
16. Irwin M, Costlow C, Williams H, Artin K H, Levin M J, Hayward A R, Oxman M N 1998 Cellular immunity to varicella-zoster virus in patients with major depression. Journal of Infectious Diseases 178S: 104–8
17. Meyers C A 1999 Mood and cognitive disorders in cancer patients receiving cytokine therapy. Advances in Experimental Medicine and Biology 461: 75–81
18. Moynihan J A, Ader R 1996 Psychoneuroimmunology: Animal models of disease. Psychosomatic Medicine 58: 546–58
19. Olness K, Ader R 1992 Conditioning as an adjunct in the pharmacotherapy of lupus very the matosus. Journal of Developmental and Behavioral Pediatrics 13: 124–5
20. Rabin B S 1999 Stress, Immune Function, and Health. WileyLiss, New York
21. Rasmussen A F, Marsh J T, Brill N Q 1957 Increased susceptibility to herpes simplex in mice subjected to avoidance learning stress or restraint. Proceedings of the Society for Experimental Biology and Medicine 96: 183–9
22. Sheridan J F, Dobbs C, Brown D, Zwilling B 1994 Psychoneuroimmunology: Stress effects on pathogenesis and immunity during infection. Clinical Microbiological Review 7: 200–12