The Biological Processes in Psychological Stress


The human stress, which is influenced by socio cultural and genetic factors, has been an important concept in understanding health and disease. Our understanding of the organic aspects of the experiencing of stress by the brain and central nervous system has greatly increased. For example, the corticotropin releasing factor/paraventricular nucleas in the hypothalamus and the locus cereleus-norepinephrine nervous system in the brainstem have been identified as divisions of the central nervous system that play a central role in the stress response. Furthermore, many of the endocrinological factors that result from the stressed central nervous system have been elucidated. Enough immunological changes (as a result of the endocrinological alterations), most of which are immunosuppressive, have been elicited to propose a biomedical model of the effects of psychological stress on mental and physical health.

Fam Syst & Health 19: 291-302, 2001

The medical literature and experience of the past many years have supported the tenet that the psychological state of a person affects his feelings of well-being and therefore may affect his mode of presentation to and treatment by the medical system. We and other physicians have noted that a person's psychological state and the type and degree of psychological stress he may be undergoing affect his feelings of well-being and daily functioning (Jemmott & Locke, 1984).

However, it has been less clear until the last several decades whether a person's stress response alters biochemical and physiological factors rendering him more susceptible to organic disease. We believe that evidence from the fields of psychoneuroendocrinology and psychoneuroimmunology is accumulating to lend affirmative support to this concept.

The human stress response may be viewed from several levels: the sociocultural level, the individual's psychological response (affecting the way the person presents to the physician), the biochemical response of the brain and central nervous system, the resulting endocronological response, and the response at the level of the human immune system (and, perhaps, concomitantly at the tissue level). While these aspects of the stress response will be separated for purposes of this discussion, they are more properly considered tightly interrelated (Figure 1).

The purpose of this article is threefold: to summarize the more recent literature on the endocrinological and immunological aspects of the human response to stress; to propose a biomedical model of the human stress response and its potential role in illness; and to stimulate future research to further clarify this model.

A person's response to stress begins when he is in a situation he perceives as stressful or difficult. A simple example of acute stress may be when one is in danger of being attacked physically. Examples of life situations often experienced as chronically and highly stressful in American society are marital disruption, loss of a job, death of a loved one, and geographical relocation. Each person's reaction to a stressor, however, is unique and generally depends on the person's individual experiencing of the stressor, which may be influenced by past life experiences and genetic factors. Also, multiple minor stressors may accumulate to result in a mild or severe stress response. These situations then result in the subject experiencing negative thoughts about the situation or himself (Burns, 1980).

The person responding to psychological stress may present to the physician with any of a variety of resulting symptoms. Common examples of these (assuming the physical examination and laboratory data fail to reveal an explanatory organic cause) may be head and neck pain, non-cardiac chest pain, dizziness, abdominal pain, and difficulty sleeping. Often, more than one symptom is presented at a time (e.g., dizziness with fatigue). Sometimes cardiac chest pain is exacerbated by stress. Sometimes the presentation of psychological stress is atypical or concealed and requires further scrutiny to ascertain the stress-related nature of the symptom.

A 31-year old woman presented to one of the authors with a
bilateral ear pain without upper respiratory symptoms.
Otoscopic examination revealed tympanic membranes with
normal landmarks and normal ear canals. On close questioning
and further examination it appeared that the pain actually
originated anterior to the auricles (in the area of the
masseter muscles). A staff member mentioned she had noticed
the patient clenching her jaw muscles excessively. When
asked if she was under any new stresses, the patient began
crying and related some difficulties she was having with her
two toddlers at home.
On the continuum of severity of stress response, perhaps the most severe manifestations of stress are major depressive disorder and panic disorder, which may be viewed as physiological mechanisms for coping with adversity which have become dysregulated. Symptoms of major depressive disorder include sadness, anhedonia, significant weight loss, changes in sleep patterns, psychomotor agitation or retardation, fatigue, inappropriate guilt, difficulty concentrating, and suicidal ideation. Symptoms of panic disorder include recurrent unexpected panic attacks (typical symptoms of which include palpitations, sweating, trembling, sensations of shortness of breath, lightheadedness, and paresthesias), persistent concern about having additional attacks, worrying about the implications of the attack, and significant change in behavior related to the attacks (American Psychiatric Association, 1994). Symptomatology often overlaps between major depressive disorder and panic disorder in clinical situations. Both illnesses carry the risk of suicide as a result of feelings of hopelessness and personal suffering.

Central Nervous System Neurochemistry of The Psychological Response To Stress
Enough research and integration has been done to propose a reasonable model of the neurochemistry of the stress response (Chrousos & Gold, 1992; McEwen, 1995). The two principle components of the central nervous system which determine stress response are the corticotropin releasing factor/paraventricular nucleas (CRF/PVN) and locus cereleus-norepinephrine/sympathetic (LC-NE) nervous systems (Chrousos & Gold, 1992). Components of the CRF/PVN system, thought to be an extremely important coordinator of the stress response, are widespread throughout the brain but appear to be most centralized in the paraventricular nucleas of the hypothalamus. High central nervous system levels of CRF activate the sympathetic nervous system and pituitary-adrenal axis causing increases in heart rate, blood pressure, and serum glucose and promoting in the individual arousal as well as cautious restraint. Higher levels of CRF produce symptoms of even more severe anxiety. CRF causes the basophilic cells of the anterior pituitary gland to induce proopiomelanocortin, a polyprotein which is cleaved to form adrenocorticotropic hormone (ACTH) and beta-endorphin. Opiates function to relieve pain during a fight-or-flight stress situation (Black, 1994).

The LC-NE/sympathetic system is also located in the brainstem. Activation of this system leads to release of NE from brain neurons resulting in enhanced arousal and anxiety in the individual (Chrousos & Gold, 1992).

Activation of either the CRF/PVN or LC-NE/sympathetic system functionally activates the other. For example, in addition to the existence of various neuroanatomical connections between the two systems, serotonin and acetylcholine appear to excite CRF/PVN neurons and the LC-NE/sympathetic system, while both of these centers are inhibited by glucocorticoids and gabaergic and opioid peptidergic neurotransmission (Chrousos & Gold, 1992). Too little of the neurotransmitter serotonin (5-hydroxytryptamine) (5-HT) is characterized in depression, and research focusing on the receptor binding site density in the brains of suicide victims found increases in both 5-HT2 and alpha2 receptor density, suggesting these systems are "upregulated" in depression. Low to moderate levels of adrenal steroids are thought to protect the brain from excess anxiety.

The stress response is considered adaptive when it is in reaction to an acute stressor and is of limited duration. In this situation the accompanying anti-anabolic, catabolic, and, as will be seen, immunosuppressive effects of this response would be beneficial, temporary, and have no detrimental consequences. Major depression, characterized by self-deprecatory and negative thinking, hypervigilance and insomnia instead of just vigilance, and dysphoric hyperarousal instead of just arousal may be thought of as dysregulation of the stress response, when the system is not brought under control by the usual regulatory mechanisms. The individual becomes less able to focus on everyday problems and learning issues. Major depression, with its frequently noted elevation in serum cortisol levels, may be caused by chronic and increased CRF secretion ("CRF overdrive") (Chrousos & Gold, 1992; Black, 1994). (Unfortunately, the elevation in serum cortisol levels frequently accompanying depression has been shown to lack enough specificity to serve as a diagnostic tool for major depression [Jefferson and Greist, 1994].) Furthermore, it is felt that panic disorder, characterized by anticipatory fear (Burns, 1980) and resembling sympathetic system discharges, is another manifestation of prolonged hyperactivity of the stress system (Chrousos & Gold, 1992). Less severe stress-related symptoms without obvious physical manifestations such as those mentioned previously which are frequently seen by primary care physicians may represent lesser degrees of hyperarousal of the stress system in which the regulators are not completely overwhelmed.

Endocrinological Factors
In this section, we will examine more recent evidence that the human body has a neuroendocrinological response to stress. Mostly prospective studies with the most powerful designs will be reviewed.

Serum levels of four hormones are commonly measured in assessing the endocrinological response to stress of an individual. These are epinephrine (a beta-1/beta-2 adrenergic receptor agonist), norepinephrine (a beta-1 and alpha adrenergic agonist), cortisol and ACTH. Changes in plasma levels of catecholamines reflect activation of the sympathetic-adrenal-medullary axis (SAM), and changes in plasma levels of cortisol reflect activation of the hypothalamus-pituitary-adrenal axis. Other hormone levels that are sometimes measured in determining the systemic endocrinological response to stress are serum prolactin, platelet serotonin, growth hormone (GH), and serum testosterone. Serum assays of catecholamines and cortisol are now accurate enough that investigators no longer have to rely on 24-hour urine assays of them.

Recent Research
Arnetz, Brenner, Levi, et al. (1991) over a 2-year period followed 354 employees, 150 of whom were about to lose their jobs due to a plant shut down, 62 with insecure jobs living under the constant threat of unemployment, and 112 securely employed workers with no threat of being laid off. Participants were sampled at regular intervals (up to 7 times) for serum cholesterol, high-density lipoprotein (HDL), cholesterol, cortisol, triglycerides, urea, albumin, as well as a number of psychological variables. Psychological stress was highest and sleep quality was poorest during the anticipatory phase of the study but, interestingly, moved toward that of the controls after job loss. Systolic blood pressure and serum cholesterol tended to be higher and HDL cholesterol was consistently depressed for the unemployed. Serum cortisol levels were elevated during the anticipatory phase of unemployment, returned toward baseline during the 6 months following unemployment, and increased around the one-year mark of unemployment.

Schedlowski, Jacobs, Alker, et al. (1993) in 45 first-time parachute jumpers showed significant rises in plasma concentrations (p > .001 for both) of epinephrine and norepinephrine during the jump with subsequent decreases to baseline levels 20 minutes later. Cortisol levels reached their peak (p < .001) 20 to 30 minutes after the jump and did not reach baseline levels within one hour. Thus, both the sympathetic-adrenal-medullary and the hypothalamus-pituitary-adrenal axes were apparently activated by the jump.

Benschop, Jacobs, Sommer, et al. (1996) also did endocrinologic studies in 25 first-time parachute jumpers but, in addition, studied the affects of various medications on the endocrinological factors. In a group given alprazolam (a benzodiazepine anxiolytic) before jumping, subjects had significantly lower epinephrine concentrations than the group given no drug (placebo group) (p < .01). Plasma cortisol levels increased slightly immediately after the jump in the group pretreated with propranolol (which causes blockade of beta-adrenergic receptors) and the placebo group and then rapidly decreased. Cortisol levels did not increase in the alprazolam-treated group. Oleshansky & Myerhoff (1992) studied 17 healthy males and found that graded treadmill to exhaustion was associated with increases in both plasma norepinephrine (239 percent) and epinephrine (201 percent) immediately after completing the exercise. They also found in 19 males that a 30-minute stressful interview was associated with elevated levels of epinephrine and norepinephrine, though epinephrine levels (only) were returning to normal well before the end of the interview.

Cacioppo, Malarky, Kiecolt-Glaser, & Uchino (1995) subjected 22 older women to a brief speech stressor followed by a brief math stressor. Cardiac preejection period (PEP), thought to reflect sympathetic activation of the heart, and respiratory sinus arrythmia (RSA), thought to reflect vagal activity to the heart, were measured. The psychological stressor produced a large increase in heart rate (p < .0001). The stressor abbreviated PEP and diminished RSA consistent with the hypothesis that psychological stressors evoke reciprocal cardiac sympathetic activation and cardiac vagal withdrawal. The stressor elevated systolic blood pressure (p < .0001) and diastolic blood pressure (p < .001). The stressor also increased plasma epinephrine (p < .0001), norepinephrine (p < .0O01), and ACTH (p < .01) concentrations. Uchino, Cacioppo, Malarkey, & Glaser (1995) subjected 24 undergraduate women to a stressful mental arithmetic task. The stressor was significantly associated with increased heart rate, systolic blood pressure, and diastolic blood pressure. PEP was significantly shortened and RSA decreased, also suggesting reciprocal sympathetic activation and parasympathetic withdrawal of cardiac chronotropy. Plasma norepinephrine (p < .001) and epinephrine (p < .05) were elevated in response to the stressor. Heart rate reactivity was significantly correlated with plasma ACTH (p < .02) and cortisol changes (p < .01). Furthermore, PEP reactivity significantly correlated with cortisol reactivity (p < .05).

This group of studies together with other non-cited studies provide evidence that in the human being during acute stress the sympathetic-adrenal-medullary axis is activated causing elevated plasma epinephrine and norepinephrine levels, and the hypothalamus-adrenal-medullary axis is activated, ultimately causing elevated plasma levels of cortisol. There also appears to be evidence that under long-term stress (without including depression as a factor), plasma cortisol levels may remain chronically elevated (Arnetz et al., 1991).

Other hormones have been shown to be affected by stress as more sensitive assaying techniques have been developed. Beta-endorphin, enkephalins, melanocyte stimulating hormone, thyroxin, thyro-tropin, vasopressin, aldosterone, renin, erythropoietin, growth hormone, insulin, glucagon, prolactin, parathyroid hormone, calcitonin, and gastrin have all been shown to increase during stress, though the exact significance of these elevations has not been worked out in all cases (Jemmott & Locke, 1984; Borysenko & Borysenko, 1982). Receptors for both met-enkephalin and beta-endorphin have been found on the lymphocyte and both have been shown to have immune-enhancing effects (Jemmott & Locke, 1984; Wybran, Appleboom, & Govaerts, 1979; Hazum, Chang, & Cuatrecasas, 1979; Gilman, Schwartz, Milner, et al., 1982).

In the following section we will consider in further detail how these endocrinological factors interact with immunologic factors in the psychologically stressed state.

Classic Studies
In 1975, Bartrop, Luckhurst, Lazarus, et al. prospectively studied 26 bereaved spouses with matched non-bereaved controls. Blood samples were taken about 2 weeks after bereavement and again 6 weeks later. Lymphocyte response to the mitogen phytohemagglutinin (PHA) was significantly depressed on the second drawing (p < .05), as was the response to the mitogen concanavalin A (Con A) (p < .05).

Demonstrating that the immune system may be sensitive to even milder stressors than bereavement, in 1984, Kiecolt-Glaser, Garner, Speicher, & Penn showed a significant (p < .003) decrease in natural killer (NK) cell activity in 75 first-year medical students during final examinations compared to a blood sample from them six weeks before examinations. Those students who scored high in the Brief Symptom Inventory for stressful life events (p < .006) also had lower NK activity than low scorers. The following year, using a similar model with 40 second-year medical students, Glaser, Kiecolt-Glaser, Stout, & Tarr showed significant decreases in total percentages of T-lymphocytes (OKT3+) (p < .01) and helper cells (OKT4+) (p < .05) in the students during final examinations than 6 weeks previous to the exams. The helper-suppressor lymphocyte ratio did not change significantly. Furthermore, they found that the T-lymphocyte response to stimulation by Con A and PHA was significantly lower during examinations, as was the lymphocyte response to PHA. In 1986, Kiecolt-Glaser, Glaser, Strain, & Stout again used their model to study 34 first-year medical students undergoing final examinations. This time, however, half of the students were randomly assigned to a hypnotic/relaxation group which met regularly before the second blood draw. The percentage of the helper/inducer T-lymphocytes (CD4+) decreased significantly in the examination sample over the sample whose blood was drawn a month before (p < .003), but the percentage of suppressor/cytotoxic (CD8+) cells did not change significantly. The helper-suppressor cell ratio (CD4+/CD8+), however, decreased significantly from the first to the second sample. Moreover, there was a significant reduction in natural killer cell activity during examinations as measured by NK cell lysis of target cells (p < .003). Interestingly, more frequent practice of relaxation was associated with higher helper/inducer cell percentages in the second blood sample.

In 1986, Glaser, Rice, Speicher, et al., again, used the model (in 40 second-year medical students) to show that the production of interferon (IF) by Con A stimulated lymphocytes declined sharply from the first to the second sample (p < .0001). There was also a significant decrease in NK activity from the first to the second sample. IF is a major regulator of NK activity because it can affect growth and differentiation of NK cells from their progenitor cells. With lower levels of IF there will be less growth of NK cells, and this may be interpreted as another facet of immunosuppression. IF can also activate the lytic activity of target-binding cells, enhance cytolysis of target cells, and increase the number of target cells that can be killed by an effector cell (Herberman, 1982).

Recent Research
In a more recent study on acute stress immunology, Bachen, Manuck, Marsland, et al. (1992) measured lymphocyte populations and PHA-stimulated T-cell mitogenesis in 33 healthy young men before and immediately after their performance of a frustrating laboratory task. Measurements after the task showed a significant reduction in T-cell mitogenesis and reduced T-helper/T-suppressor cell ratio and elevation in NK cells.

In the aforementioned study by Arnetz et al. (1991), PHA reactivity of lymphocytes decreased (p < .05) among their unemployed subjects after 9 to 12 months of unemployment but returned to normal at 20 months.

In the study by Schedlowski et al. (1993), in addition to elevations of catecholemines and cortisol during the parachute jump, there were significant increases in CD2+, CD3+, CD4+, and CD8+ cells immediately after the jump and an even more pronounced rise in NK cells.

In the study by Benshop, Jacobs & Sommer (1996) on first time parachute jumpers, NK cell numbers and activity increased during the jump but these changes were inhibited by propranolol. Pretreatment with alprazolam also inhibited increase in NK cell activity. They hypothesize that these increases in circulating lymphocytes are a result of mobilization of lymphocytes from the spleen.

In the study by Cacioppo et al. (1995) of 22 women subjected to a brief psychological stressor, in addition to the previously discussed endocrinological changes, it was noted that T-lymphocyte and NK cell proliferation to Con A decreased as a function of concentration (p < .001) and of the psychological stressor (p < .05), the blastogenic response to PHA decreased as a function of concentration (p < .01) but did not decrease significantly as a function of the stressor (p > .10), and NK cell cytotoxicity increased after the psychological stressor. In addition, the stressor decreased the percentage of CD4+ cells (p < .001), increased the percentage of CD8+ cells (p < .05), reduced the CD4+/CD8+ ratio (p < .001), and increased the percentage of NK cells (p < .005).

In 1994, Zorrilla, Redei, & DeRubeis studied the relationship of state distress to T-cell function and cytokine levels in 40 male college freshman. They found that subjects who were characteristically more anxious had more anxious mood and significantly lower lymphocyte proliferative responses to Con A as well as lower levels of circulating interleukin- 1Beta (IL- 1Beta). Subjects with more negative attributional styles for bad events had reduced Con-A stimulated T-cell responses and lower levels of circulating IL-2. The same year Schulz & Schulz studied 37 medical students undergoing oral examinations and compared them with 32 students not undergoing examinations as controls. They found that plasma concentrations of IL-1Beta were decreased in the exam group at the time of stress compared to a stress-free time 6 weeks later (p < .001). No such difference was seen for the control group (p = .52). Furthermore, plasma levels of IL-2R were significantly lower than those of the control group at the time of the exam (p < .001). In a second study, 26 athletes were studied while undergoing exhaustive short term physical stress and psychological stress. Plasma concentrations of IL-1RA, IL-6, IL-6R, and of the pituitary hormones ACTH, GH, and prolactin, increased after strenuous exercise, whereas IL-1, IL-2R, and ICAM-1 did not. IL-IRA decreased under psychological stress while IL-6R and the hormones ACTH and prolactin increased.

Shimizu, Kawamura, Miyaji, & Oya (2000) exposed mice to restraint stress and noticed that severe lymphopenia was induced in all immune system organs. This response, however, was completely absent in adrenalectomized mice (which would have no endogenous source of corticosteroid but could still produce epinephrine through the remaining sympathetic nervous system). CD4+ and NK cells were resistant to stress in the latter mice. It was therefore felt that endogenous steroid hormones were important in the induction of immunosuppressive states after stress.

Seminov, Vargin, Ozherelkov, & Semenova (1998) also exposed mice to restraint stress and noticed a 3 to 8-fold increase in the number of splenic macrophages. Simultaneous with this, the stressed mice also exhibited an increase in sensitivity to intraperitoneal infection with Langat virus, which they interpreted as a sign of immunosuppression.

Are plasma elevations of catecholamines directly responsible for changes in immune functioning? A few studies shed light on this issue. For example, Crary, Borysenko, Sutherland, et al. (1983) injected 0.2 mg of epinephrine subcutaneously in six healthy adult volunteers and drew blood samples at various times afterward. Mononuclear cells isolated from these blood draws were cultured in the presence of mitogens. The blastogenic responses to pokeweed mitogen and PHA were significantly reduced (p < .05) for up to 60 minutes after the injection. The response to Con A was reduced in the 15-minute samples only. Adrenergic receptors have been shown to be present on lymphocytes, and catecholamines have been shown to modulate various immune effector functions (Crary et al., 1993; Williams, Snyder, & Leftowitz, 1976; Besedovsky, del Ray, Sorkin et al., 1979). Stimulation of beta-adrenergic (epi-nephrine) receptors on T-lymphocytes, B-lymphocytes, and macrophages decreases cellular response. These effects are apparently mediated by 3'5'adenosine monophosphate (cyclic AMP) (Jemmott & Locke, 1984; Henny, Bourne, & Lichtenstein, 1972).

What is the mechanism by which corticosteroid affect immunity? The effects of corticosteroid on the immune system are complex, but in general they are antiinflammatory and immunosuppressive (Jemmott & Locke, 1984; Claman, 1992; Fauci, 1978). Increased serum cortisol has been associated with decreased lymphocyte response to mitogen and a decreased ability of lymphocytes to destroy foreign cells (Jemmott & Locke, 1984; Claman, 1972).

Studies which are able to show damage to tissues directly from psychological stress are rare. There are a few studies which have shown correlation between illness rates and stress. For example, Cohen, Tyrrell, & Smith (1991) administered nasal drops containing one of five respiratory viruses to 394 healthy subjects and saline nose drops to an additional 26 after giving them questionnaires assessing their degrees of psychological stress. The rates of respiratory infection (p < .005) and clinical colds (p < .02) increased in a dose-response fashion with increases in psychological stress. Analysis of various stress-illness mediators and controls for personality variables failed to alter the findings.

Figure 2 is a model proposed to represent the acutely stressed state within the person. The same or similar model is proposed to represent the chronically stressed human being. The person is in a situation he perceives, as processed by the cerebral cortex, as stressful or threatening. This activates the hypothalamus-pituitary-adrenal axis by stimulating the CRF/PVN system, which interacts with the LC-NE system. The CRF/PVN releases excess CRF to the pituitary which releases excess adrenocorticotrophic hormone to the bloodstream. This stimulates the adrenal cortex to release excess corticosteroid to the bloodstream. The anti-inflammatory and immuno-suppressive properties of the corticosteroid cause a decreased T-lymphocyte response to stimulated mitosis and therefore decreased production of IL-2, IL-1Beta, IL-2R, interferon and other interleukins.

The LC-NE, as above, is likewise stimulated, interacting with the CRF/PVN and causing activation of the sympathetic-adrenal-medullary axis. The sympathetic nervous system causes increased discharging of catecholamines from the adrenal medulla as well as increased discharging of catecholamines from the SAM itself. Increased catecholamine serum levels also contribute to decreased stimulated mitosis of T-lymphocytes, presumably through direct interaction of catecholamines with binding sites on lymphocytes, and therefore decreases in some interleukins and interferon. For similar reasons, increased serum catecholamine levels cause an increase in NK cells, increase in NK cell activity, relative decreases in CD4+ counts, increases in CD8+ counts, and a decreased CD4+/ratio.

This response to stress is reversed when the person is no longer in the stressful situation or if he cognitively perceives that the situation is no longer threatening, because signals signifying danger are no longer being relayed from the cerebrum to the hypothalamus, where it is thought the biological response to stress is initiated. In addition, relatively high levels of intrinsic corticosteroid can feed back to the anterior pituitary gland causing it to decrease production of ACTH (Biller & Daniels, 1998). If the person cannot remove himself from the situation that he perceives as stressful or if he is unable to cope with the situation because of a dysfunctional psychological process, the stress response becomes chronic, and he is made prone to anxiety, depression, or perhaps some other mental disorder.

There probably are subtle complexities and other interactions yet to be studied in this model. Further research will be needed in order to elicit all facets of the person's psychosocial and biochemical response to psychological stress. Future research should concentrate on fully eliciting the chemical interactions between the stress hormones and T-cells (resulting generally in their immunosuppression). Future research should also concentrate on the cytokines, since the responses of all these chemicals have not been examined in the psychologically stressed state. Further understanding of the biopsychosocial response to stress will lead to improved sociocultural, psychotherapeutic, pharmacologic, and other approaches with which we may intervene in stress-related disorders.

DIAGRAM: Figure 1. A general schema of the cascade of the biopsychosocial model of psychological stress from the primary care viewpoint.

DIAGRAM: Figure 2 Diagram of the biomedical events leading to psychological stress response.

REFERENCES American Psychiatric Association. (1994). Diagnostic and Statistical Manual of Mental Disorders. (4th ed.). Washington, DC, American Psychiatric Association.

Arnetz, B.B., Brenner, S-O., Levi, L., Hjelm, R., Peterson, I-L., Wasserman, J., Petrini, B., Eneroth, P., Kallner, A., Kvetnansky, R., & Vigus, M. (1991). Neuroendocrine and immunologic effects of unemployment and job insecurity. Psychotherapy and Psychosomatics, 55, 76-80.

Bachen, E.A., Manuch, S.B., Marsland, A.L., Cohen, S.D., Malkoff, S.B., Muldoon, M.F., & Rabin, B.S. (1992). Lymphocyte subset and cellular immune responses to a brief experimental stressor. Psychosomatic Medicine, 54, 673-679.

Bartrop, R.W., Luckhurst, E., Lazarus, L., Kiloh, L.G., & Penny, R. (1977). Depressed lymphocyte function after bereavement. Lancet, 1, 834-836.

Besedovsky, H., del Ray, A., Klusman, I., Furukawa H., Monge Arditi, G., & Kabiersch, A. (1991). Cytokines as modulators of the hypothalamus-pituitary-adrenal axis. Journal of Steroid Biochemistry and Molecular Biology, 40, 613-618.

Benschop, R.J., Jacobs, R., & Sommers, B. (1996). Modulation of the immunologic response to acute stress in humans by beta-blockade or benzodiazepines. The FASEB Journal, 10, 517-254.

Biller, B.M.K., & Daniels, G.H. (1998). Neuroendocrine regulation and diseases of the anterior pituitary and hypothalamus. In A.S. Fauci, E. Braunwald, K.L. Isselbacher, et al. (Eds.), Harrison's Principles of Internal Medicine. (14th ed., pp. 1985-1986). New York, NY: McGraw-Hill Companies, Inc.

Black, P.H. (1994). Central nervous system-immune system interactions: psychoneuroimmunology of stress and its immune consequences. Antimicrobial Agents Chemotherapy, 38, 1-6.

Borysenko, M., & Borysenko, J. (1982). Stress, behavior, and immunity: animal models and mediating mechanisms. General Hospital Psychiatry, 4, 59-67.

Burns D.D. (1980). Feeling good: The new mood therapy. New York, NY: William Morrow and Company, Inc.

Cacioppo, J.T., Malarky, W.B., Kiecolt-Glaser, J.K., Uchino, B.N., Sgoutas-Emch, S.A., Sheridan, J.F., Berntson, G.G., & Glaser, R. (1995). Heterogeneity in neuroendocrine and immune responses in brief psychological stressors as a function of autonomic cardiac activation. Psychosomatic Medicine, 57, 154-164.

Chrousos, G.P., & Gold, P.W. (1992). The concepts of stress and stress system disorders: overview of physical and behavioral homeostasis. Journal of the American Medical Association, 267, 1244-1252.

Claman, H.N. (1972). Corticosteroid and lymphoid cells. The New England Journal of Medicine, 287, 388-397.

Cohen, S., Tyrrell, D.A.J., Smith, A.P. (1991). Psychological stress and susceptibility to the common cold. The New England Journal of Medicine, 325, 606-611.

Crary, B., Borysenko, M., Sutherland, D.C., Kutz, I., Borysenko, J.Z., & Benson, H. (1993). Decrease in mitogen responsiveness of mononuclear cells from peripheral blood after epinephrine administration in humans. Journal of Immunology, 130, 694-697.

Fauci, A.S. (1978). Mechanisms of the immunosuppressive and anti-inflammatory effects of glucocorticoids. Journal of Immunopharmacology, 9, 1-25.

Gilman, S.C., Schwartz, J.M., Milner, R.J., Bloom, F.E., & Feldman, J.D. (1982). Beta-endorphin enhances lymphocyte proliferative responses. Proceedings of the National Academy of Sciences, USA 79, 4225-4230.

Glaser, R., Kiecolt-Glaser, J.K., Stout, J.C., Tarr, K.L., Speicher, C.E., & Holliday, J. (1985). Stress-related impairments in cellular immunity. Psychiatry Research, 16, 233-239.

Glaser, R., Rice, J., Speicher, C.E., Stout, J.C., & Kiecolt-Glaser, J.K. (1996). Stress depresses interferon production by leukocytes concomitant with a decrease in natural killer cell activity. Behavioral Neuroscience, 100, 675-678.

Hazum, E., Chang, K.J., & Cuatrecasas, P. (1979). Specific nonopiate receptors for beta-endorphin. Science, 205, 1033-1035.

Henny, C.S., Bourne, H.R., & Lichtenstein, L.M. (1972). The role of cyclic 3'-5' adenosine monophosphate in the specific cytolytic activity of lymphocytes. Journal of Immunology, 108, 1526-1534.

Herberman R.B. (1982). Possible effects of central nervous system on natural killer (NK) cell activity. In S.M. Levy (Ed.), Biological Mediators of Health and Disease: Neoplasia (pp.235-248). New York, NY: Elsevier.

Jefferson, J.W., & Greist, J.H. (1994). Mood Disorders. In R. E. Hales, S.C. Yudovsky, J.A. Talbot. The American Psychiatric Press Textbook of Psychiatry. (2nd ed., pp. 472-476). Washington, DC: American Psychiatric Press.

Jemmott, III, J.B., & Locke, S.E. (1984). Psychosocial factors, immunologic mediation, and human susceptibility to infectious diseases: how much do we know? Psychological Bulletin, 1, 78-108.

Kiecolt-Glaser, J.K., Garner, W., Speicher, C., Penn, G.M., Holliday, J., & Glaser, R. (1984). Psychosocial modifiers of immunocompetence in medical students. Psychosomatic Medicine, 1, 7-14.

Kiecolt-Glaser, J.K., Glaser, R., Strain, E.C., Stout, J.C., Tarr, K.L., Holliday, J., & Speicher, C.C. (1986). Modulation of cellular immunity in medical students. Journal of Behavioral Medicine, 9, 5-21.

McEwen, B.S. (1995). Stressful experience, brain, and emotions: developmental, genetic, and hormonal influences. In M. S. Gazzanga, The Cognitive Neurosciences (pp. 1117-1138). Cambridge, MA: MIT Press.

Oleshansky, M.A., & Meyerhoff, J.L. (1992). Acute catecholaminergic responses to mental and physical stressors in man. Stress Medicine, 8, 175-179.

Schedlowski, M., Jacobs, R., Alker J., Prohl, F., Stratmann, G., Richter, S., Hadicke, A., Wagner, T.O.F., Schmidt, R.E., & Tewes, U. (1995). Psychophysiological, neuroendocrine and cellular immune reactions under psychological stress. Neuropsychobiology, 28, 87-90.

Schulz, H., & Schulz, K-H. (1994). Effects of psychological stress and exercise on different parameters of cytokines. Psychologische Beitrage Band, 36, 36-46.

Semenov, B.F., Vargin, V.V., Ozherelkov, S.V., & Semenova, T.B. (1998). Stress increases the population of splenic macrophages, permissive for Langat virus in mice. Zhurnal Mikrobiologii, Epidemiologii, i Immunobiologii, 3, 57-60.

Shimizu, T., Kwamuro, T., Miyaji, C., Oya, H., Bannai, M., Yamamoto, S., Weerasinghe, A., Halder, R.C., Watanabe, H., Hatakeyama, K., & Abo, T. (2000). Resistance ofextrathmic T cells to stress and the role of endogenous glucocorticoids in stress associated immunosuppression. Scandanavian Journal of Immunology, 51,285-292.

Uchino, B.N., Cacioppo, J.T., Malarky, W., & Glaser, R. (1995). Individual differences in cardiac sympathetic control predict endocrine and immune responses to acute psychological stress. Journal of Personality and Social Psychology, 69, 736-743. Williams, L.T., Snyderman, R., Lefkowitz, R.J. (1976). Identification of beta-adrenergic receptors in human lymphocytes by (-)3H]alprenolol binding. Journal of Clinical Investigation, 57, 149-155.

Wybran, J., Appelboom, T., Famaey, J.P., & Govaerts, A. (1979). Suggestive evidence for receptors for morphine and methionine-enkephalin on normal human T lymphocytes. Journal of Immunology, 123, 1068-1070.

Zorrilla, E.P., Redei, E., DeRubeis, R.J. (1994). Reduced cytokine levels and T-cell function in healthy males: relation to individual differences in subclinical anxiety. Brain Behavior and Immunology, 8, 293-312.


By Richard I. Haddy, M.D. and Richard D. Clover, M.D.

Richard I. Haddy, M.D., Department of Family and Community Medicine, University of Louisville School of Medicine, Louisville, Kentucky 40292; 501-852-1250;

Richard D. Clover, M.D., Department of Family and Community Medicine, University of Louisville School of Medicine, Louisville, Kentucky 40292.

Share this with your friends