Implications of Stress, Psychosocial Factors on the Immune System

In the 1950s and 1960s, researchers found that mice and rats subjected to stressful stimuli were more likely to develop viral infections and tumors than nonstressed animals (Miller, 1998). Today, that once-pioneering research in psychoneuroimmunology (PNI) has burgeoned into sophisticated clinical studies that look at, for example, how caregiving can affect the immune system, how stress may delay wound healing and how pretreatment with an antidepressant prevents cytokine-induced depression in therapy for cancer.

In an interview with Psychiatric Times, Ronald Glaser, Ph.D., professor of molecular virology, immunology and medical genetics at Ohio State University College of Medicine, defined psychoneuroimmunology as a "field that studies the interactions between the central nervous system, the endocrine system and the immune system; the impact of behavior/stress on these interactions; and the implications for health of these interactions." It also examines how psychological and pharmacological interventions may modulate these interactions (Glaser et al., 1999b).

Current research emphasizes bidirectional communication between the nervous and immune systems, according to Adrian J. Dunn, Ph.D., professor and head of the department of pharmacology and therapeutics at Louisiana State University Medical Center.

"The major form of communication is chemical messengers, whether they are secreted by nerve cells, endocrine organs or immune cells," Dunn said. "The nervous system is known to influence immune system function via neurotransmitters released by nerve terminals and indirectly via secretions from endocrine organs. In turn, the immune system is known to influence nervous system function by cytokines and other peptides secreted by immune cells. Other mechanisms may exist, but they have not yet been identified" (Dunn, 1995).

The many aspects of psychoneuroimmunology, from molecular biology to behavioral and clinical studies, are being investigated at Ohio State University College of Medicine by Glaser; by psychologist Janice Kiecolt-Glaser, Ph.D., Glaser's wife and primary collaborator; and by 11 other faculty members. The Glasers have been working in the field since 1982 "when the field was relatively young."

At the time, there were only three human studies in psychoneuroimmunology, Glaser told PT, and those studies examined very severe stressors, such as splashdown for astronauts, or exposure to 24 hours of noise, or bereavement of a spouse. No studies had been conducted on everyday stressors.

"So Jan and I talked about what to do and agreed to do PNI studies on young healthy adults [medical students], just to learn how stress could affect the immune system. For 10 years we worked with that academic stress model, and it was very productive for us in learning about stress and immunity," Glaser said.

Since then, the Glasers and their group have studied the effects of a number of psychosocial factors on immune function, including marital conflict (Kiecolt-Glaser et al., 1997) and the stress of caring for family members with Alzheimer's disease (Glaser et al., 1998). They also have studied the health implications of psychological stressors.

"The first study that we did that would fall in this category of health implications...was a study that showed that academic stress and anxiety could affect how a person responds to a hepatitis B vaccination," Glaser said.

Forty-eight medical students were inoculated with a series of three injections of the hepatitis B vaccine to coincide with the third day of three, three-day examination blocks. Twelve of the 48 seroconverted after the first injection; these students were characterized by falling into the lower stressed/lower anxiety group of students. Students who reported greater social support and lower anxiety and stress demonstrated a higher antibody response to the vaccine and a more vigorous T-cell response to the hepatitis B surface antigen at the end of the third examination experience (Glaser et al., 1992; Glaser et al., 1998).

In another study on the influence of psychological stress on the immune response to vaccines, Kiecolt-Glaser et al. (1996) looked at spousal caregivers for Alzheimer's disease patients. The average age of the caregivers was 70 years.

In that study "we have really two things going on," Glaser said. "First, the immune systems of the caregivers were generally less vigorous because they were older, and second, they were experiencing a long-term, chronic stressor-that of caregiving for an Alzheimer's disease patient. The control group participants were very well-matched human subjects, matched in age and so forth, but not involved in caregiving situations.

"When we compared all the immune measures we looked at between the two groups, we found that the caregivers had negative changes in their immune response compared to the controls, and they also had more colds."

The researchers then administered an influenza virus vaccination to 32 caregivers and matched control subjects.

"Influenza is important for this group because influenza and the side effects of influenza, pneumonia...rank fourth in morbidity for people 75 years of age and older," Glaser said. "We found that caregiving stress resulted in a significantly dampened antibody response to the flu vaccine."

Additionally, caregivers also had lower levels of in vitro virus-specific-induced interleukin-2 levels and interleukin-1 ?. The results, Glaser said, have "significant health implications for risks of getting the flu in a population that cannot afford to get the flu because of the other risk factors that go along with that."

Research also is being conducted on caregivers of Alzheimer's disease patients at the University of New Mexico in Albuquerque. Sharon Lewis, R.N., Ph.D., professor of nursing and research associate professor in the department of pathology, and colleagues have been comparing natural killer (NK) cell number and cytotoxicity and psychosocial variables in both spousal and children caregivers of patients in the early and middle stages of Alzheimer's disease. Nineteen spousal caregivers and 13 children caregivers completed questionnaires assessing physical, mental, social and emotional health. Blood was obtained to determine NK cell number and cytotoxicity using flow cytometry (Lewis et al., 1999).

"What we found was that, as a whole, the children caregivers are much more stressed, they are much more depressed, and they have a lower quality of life; the differences are profound," Lewis said. She added that the spousal caregivers had higher NK cell numbers and NK cytotoxicity than the children caregivers on the immune parameters.

As part of the study, both groups of caregivers received an intervention in which they were asked to listen to a 30-minute relaxation tape each day and practice abdominal breathing.

"When they undergo the relaxation intervention, there is definite improvement in both groups, but the spousal caregivers do better than the children caregivers," Lewis added.

Wound Healing

Impairment of wound healing is a well-recognized sequela of diabetes, advanced age and other conditions that alter immune function. In the last few years, Glaser, Kiecolt-Glaser, and their colleagues Phillip Marucha, D.M.D., Ph.D.; William Malarkey, M.D.; and John F. Sheridan, Ph.D., have been focusing on psychological stress and wound healing. They have published four separate papers on the subject, the most recent one in Archives of General Psychiatry (Glaser et al., 1999a). In human subjects and in animal models, the group has found that stress markedly delays wound healing.

In the most recent study (Glaser et al., 1999a), skin blisters were induced by suction on the forearms of 36 postmenopausal women (mean age, 57 years). After the blister roofs were removed, a plastic template was taped to the arms and wells were filled with 70% autologous serum in buffer. Specimens were aspirated from the blister chamber wells five and 24 hours after wounding.

Women with higher perceived stress scores demonstrated significantly lower levels of two proinflammatory cytokines important for the early stages of wound healing-interleukin-1 ? and interleukin-8-at the wound site. Additionally, the women who had low levels of both cytokines after 24 hours reported more stress and negative affect, and they had higher levels of salivary cortisol than those who had high cytokine levels.

These findings, Glaser said, have "major implications for people getting surgery. So for now and the foreseeable future, we will be looking at how stress affects the wound healing process and whether a behavioral intervention, a stress reduction intervention, could buffer some of that."

Examples of such interventions would be stress reduction groups, Glaser said, "where you train people to do relaxation, guided imagery and muscle relaxation exercises. These have been shown to decrease heart rate and blood pressure."

Pharmacological Interventions

While most of the interventions are behavioral, there is at least one pharmacological intervention being explored. Andrew H. Miller, M.D., associate professor of psychiatry and behavioral sciences at Emory University School of Medicine, and colleagues recently conducted a double-blind, placebo-controlled trial in which they pretreated patients with either placebo or paroxetine (Paxil) for two weeks prior to high-dose interferon ? treatment for malignant melanoma.

Interferon ? is used to treat viral-like disorders, hepatitis C and cancers. Patients receiving high doses of interferon ? develop what French researcher Robert Dantzer, D.V.M., Ph.D., first described as sickness behavior in animals (Dantzer et al., 1998).

"The sickness syndrome that has been described includes symptoms of anhedonia, malaise, weakness, fatigue, disruption of sleep patterns [and] anorexia," Miller said. "It has intrigued many of us that these symptoms that one sees with sickness behavior overlap with those symptoms apparent in patients who have depression. Therefore, there has been a lot of interest in whether or not drugs that have been used to treat depression could be used to treat this sickness syndrome."

Miller said that a few clinicians have used antidepressants to treat some patients who developed this sickness behavior while taking cytokine therapies, and they have achieved "some success." Miller and his colleagues were particularly intrigued by a study that was done on laboratory rats. An Israeli researcher pretreated the rats for three weeks with the antidepressant imipramine (Tofranil), and then gave the rats injections of lipopolysaccharide (LPS), a cytokine inducer (Yirmiya, 1996).

"[LPS] is the factor that causes endotoxic shock in people who have serious bacterial infections. And generally, if you give this LPS to rats, they develop sickness behavior," Miller said. Reduction in the rats' preference for and consumption of saccharin solutions was used as a model of one essential feature of depression-anhedonia. Yirmiya found that chronic treatment with imipramine abolished the suppressive effect of LPS on the rats' saccharin preference and also on food consumption and social interaction.

"What we wanted to know is whether we could do a similar thing in humans," Miller said. "We enrolled 21 patients into the study [Miller et al., 1999], and 19 were able to complete the study as we originally designed it. Of those 19, 10 were on placebo and nine on paroxetine. The study ran for 14 weeks. We evaluated the patients weekly for the first four weeks, and monthly for the remaining two months of the study.

"What we found is that...90% of the placebo patients became depressed. In the paroxetine group, however, only two patients out of nine became depressed," he said.

According to Miller, there were significant differences between the groups on the 21-item Hamilton Depression Rating Scale scores with the paroxetine-treated patients showing much lower levels of depression, especially during the last two months of the study, compared to the placebo-treated group. Patients in the paroxetine-treated group also exhibited fewer symptoms on the Neurotoxicity Rating Scale that measures not only depression, anxiety, memory problems, withdrawal and isolation, but also muscle aches and pains, nausea, vomiting, vision changes, and other physical complaints.

"In fact, there was no increase in those symptoms across the course of the study in the paroxetine-treated patients compared to the placebo-treated patients, who showed steady increase in those types of symptoms," said Miller.

Once the study ended and the patients were unblinded, those on placebo were offered the opportunity to take paroxetine. Four patients switched over, and three of the four got better. One patient discontinued the paroxetine because of jitteriness.

Asked to explain why there seemed to be so few pharmacological interventions in psychoneuroimmunology research, Miller explained, "The predominant participants in the field of psychoneuroimmunology have tended to be health psychologists and, relatively speaking, fewer [medical doctors]. Health psychologists, and psychologists in general, are going to approach manipulating the physiology through making psychological interventions. Whereas a physician is generally more interested in basic physiological mechanisms...[A physician who] finds out that stress elevates a certain hormone or certain neuropeptide that leads to changes in the immune system...might intervene at the level of the hormone or peptide by developing a medication that blocks the hormone or peptide."

Miller offered an example of how the differing theoretical frameworks influence research.

"There is a lot [of] interest in the hypothalamic-pituitary-adrenal axis and corticotropin-releasing hormone [CRH] and in CRH's role in mediating the effects of stress on the immune system," Miller said. "Whereas a health psychologist may focus on relaxation techniques and coping mechanisms or cognitive-behavioral types of approaches to reduce the amount of stress the person perceives, a [physician's] approach would be to use a CRH-receptor antagonist that interrupts the hormonal pathways that translate stress from the brain to the immune system."

Already, there are CRH-antagonists that have been identified, and research is underway to determine their safety and efficacy, Miller said, adding, "the potential is there for reducing stress effects on the immune system."

Alternatively, Miller stated, "somebody might say, well, if there is this sickness syndrome that is caused by certain cytokines, and we can figure out which cytokines cause patients to feel depressed and toxic, then we could administer antagonists to those cytokines and block the development of sickness behavior."

Miller added that he and other researchers believe that the immune system may participate in depression in medically ill patients and possibly in patients whose depressions are not directly related to cytokine administration or medical illness.

"In medically ill patients, there may be a lot of tissue destruction, damage and inflammation with the resultant release of proinflammatory cytokines, which include tumor necrosis factor [TNF], interleukin-1 [IL-1] and interleukin-6 [IL-6]. These cytokines are well-known to induce this sickness syndrome and are probably the cytokines that mediate the sickness syndrome in the patients who take interferon ?," he said.

Another area of current research, according to Miller, is the role of glucocorticoid hormones in regulating the immune response during viral infection.

"What we know is that when animals are infected with a virus, it leads to the release of glucocorticoid hormones. Our interests are 1) What are the factors being released by the immune system that stimulate the body to produce these hormones? 2) What role do these hormones have in subsequently regulating the immune response," Miller said. "What we found [is that] interleukin-6 is a key cytokine for stimulating the release of glucocorticoids during viral infection. If glucocorticoids are not there or are not working in the body, then the immune system goes crazy, and it becomes unrestrained and out-of-control, and actually can lead to death. If we remove the glucocorticoids, the animals will die of a viral infection that would normally not kill them."

Mediating that death process, Miller said, is TNF.

"TNF helps control viral replication and kills cells that are infected. But when not controlled, TNF is what makes the animals get sick and then die. If we put glucocorticoids back in, TNF comes back down. Interestingly, these animals aren't dying by being overrun by virus. In fact, the animals in which we removed the glucocorticoids actually have less virus, but more TNF," Miller said.

"We and others have demonstrated that interferon ? is a potent inducer of interleukin-6, and we think IL-6 is what causes the very profound stimulation of glucocorticoids that occurs in patients given interferon ?. So you go from interferon ? to IL-6 to big elevations in glucocorticoids in our melanoma patients. These glucocorticoids should control an excessive release of TNF. However, we think that in some patients something happens to disrupt the inhibitory action of glucocorticoids-possibly an effect of cytokines to impair the functioning of receptors of glucocorticoids. With inadequate glucocorticoid-mediated cytokine control, we think that the cytokine that is making the interferon-treated melanoma patients sick is likely to be TNF, because that is the one that makes our mice sick and die."

Currently, Miller and colleagues are taking samples from patients receiving interferon ? who are depressed and looking to see if they have elevated levels of TNF.

"There actually is a TNF-antagonist available. If we found lots of TNF, then we might consider giving TNF-antagonist as a way to treat the syndrome and possibly as a treatment in other medically ill patients who are expressing depression that we feel might be based on cytokines, but that is a long way down the road."

References:

References

1.

Dantzer R, Bluthe RM, Gheusi G et al. (1998), Molecular basis of sickness behavior. Ann N Y Acad Sci 856:132-138.

2.

Dunn AJ (1995), Psychoneuroimmunology: introduction and general perspectives. In: Stress, the Immune System and Psychiatry, Leonard B, Miller K, eds. New York: John Wiley & Sons Ltd.

3.

Glaser R, Kiecolt-Glaser JK, Bonneau RH et al. (1992), Stress-induced modulation of the immune response to recombinant hepatitis B vaccine. Psychosom Med 54(1):22-29.

4.

Glaser R, Kiecolt-Glaser JK, Malarkey WB, Sheridan JF (1998), The influence of psychological stress on the immune response to vaccines. Ann N Y Acad Sci 840:649-655.

5.

Glaser R, Kiecolt-Glaser JK, Marucha PT et al. (1999a), Stress-related changes in proinflammatory cytokine production in wounds. Arch Gen Psychiatry 56(5):450-456.

6.

Glaser R, Rabin B, Chesney M et al. (1999b), Stress-induced immunomodulation. Implications for infectious diseases? JAMA 281(24):2268-2270.

7.

Kiecolt-Glaser JK, Glaser R, Cacioppo JT et al. (1997), Marital conflict in older adults: endocrinological and immunological correlates. Psychosom Med 59(4):339-349.

8.

Kiecolt-Glaser JK, Glaser R, Gravenstein S et al. (1996), Chronic stress alters the immune response to influenza virus vaccine in older adults. Proc Natl Acad Sci U S A 93(7):3043-3047.

9.

Lewis S, Kroening C, Cooper C et al. (1999), Psychosocial and immune parameters in spousal and children caregivers of patients with Alzheimer's disease. Presented at the PsychoNeuroImmunology Research Society meeting in Galveston, Texas. Available at:

www.pnirs.org/pastmeetings/99Galv/ P1c.html

. Accessed July 11.

10.

Miller AH (1998), Neuroendocrine and immune system interactions in stress and depression. Psychiatr Clin North Am 21(2):443-463.

11.

Miller A, Musselman D, Penna S et al. (1999), Pretreatment with the antidepressant, paroxetine, prevents cytokine-induced depression during IFN-alpha therapy for malignant melanoma. Presented at the PsychoNeuroImmunology Research Society meeting in Galveston, Texas. Available at:

www.pnirs.org/pastmeetings/99Galv/S1.html

. Accessed July 11.

12.

Yirmiya R (1996), Endotoxin produces a depressive-like episode in rats. Brain Res 711(1-2):163-174.