An acute stress exposure, such as a single bout of exercise, exerts profound effects on the immune system. Although the description of these immune changes has been a prolific area of research, the immunologic significance and clinical relevance of these relatively short-term changes have yet to be determined. In this review, we provide support for the hypothesis that the increased number of circulating lymphocytes and inflammatory modulators induced by exposure to acute physical or, indeed, psychological stressors might enhance the organism's ability to respond to a contiguous immune challenge.
In the last decade, Dhabhar (7) has repeatedly shown an enhanced cell-mediated immune response when acute stress occurred immediately before initial vaccination (for review, see (7)). He hypothesized that acute stress-induced immunoenhancement is related to the rapid changes in leukocyte distribution, largely because of demargination, observed with acute stress, which provide for increased immune surveillance, and prepare the organism for potential immune challenges. Dhabhar proposed that changes in leukocyte distribution can be considered a defense mechanism to challenge, and has used military metaphors to describe the hypothesis (7). The initial increase in leukocyte number in the blood are likened to soldiers exiting the "barracks" (spleen and connective tissue), where they normally reside during inactive periods, and entering the "boulevards" (circulating blood) to patrol, ready for action. Moreover, the reduction below baseline levels in leukocyte number after acute stress exposure is hypothesized to be caused by recruitment of leukocytes to potential "battle stations" (skin and lymph nodes) in preparation for immune challenge.
This acute stress-induced immunoenhancement hypothesis resonates with the recent advocacy for endogenous adjuvants in vaccination. Vaccines have been one of the most successful medical interventions to improve human health. The development of immunologic adjuvants to improve vaccine efficacy is a critical component in the design of nonreplicating vaccines. In this regard, substantial attention has been paid to the addition of exogenous substances to vaccines. Recently, interest has also been expressed in the possible application of endogenous immune activation before vaccine administration(25). Matzinger's (19) "danger" hypothesis suggests how endogenous signals might act as adjuvants for vaccination: damaged or stressed cells release signals, such as heat shock proteins or uric acid, which activate an immune response. As such, during an infection, an immune response would begin after infected cells released such signals. Particularly relevant for vaccination is the proposal that stressed cells (e.g., damaged myocytes) may release danger signals that activate dendritic cells (14). Indeed, it has been found that damaged cells in vivo stimulated antigen-presenting dendritic cells to acquire antigen and/or to mature and migrate to lymph nodes (25). Furthermore, the in vivo injection of heat shock proteins, which are examples of Matzinger's proposed danger signals, causes a large increase in dendritic cells in the draining lymph nodes (3). It is through these mechanisms that activation of the immune response by endogenous signals may enhance the response to vaccination. Below, we put forward the case that acute exercise, which initiates the release of these endogenous danger signals, can thereby act as an adjuvant to vaccination.
Early evidence for acute stress-induced immunoenhancement comes from animal research. It can be divided into those studies examining humoral (antibody) responses to vaccination and those using cell-mediated responses to antigen presentation. Rodent studies of acute stress exposure in close temporal proximity to vaccination provide support for its immunoenhancing effect on antibody responses. Studies have shown that exposure to foot shock before vaccination with the novel protein keyhole limpet hemocyanin (KLH) was associated with enhanced antibody responses relative to nonexposed controls (e.g., (20,30)). Restraint stress has also been demonstrated to enhance response to vaccination, with the importance of the acute nature of stress highlighted. Rats exposed to brief (2 h × 2 d) stress exhibited higher antibody titers after sheep red blood cell (SRBC) vaccination than control rats; in contrast, rats exposed to extended stress (6 h × 4 d) showed no difference to control rats (20). Similarly, in mice, it recently has been shown that 2 h of restraint stress immediately before vaccination with SRBC enhanced the antibody response; conversely, animals that had been exposed for 6 wk of chronic mild stress, but were left undisturbed for 18 h before immunization, exhibited lower antibody responses than control animals (27).
Animal studies have often used cell-mediated responses to vaccination as the primary outcome measure. The delayed-type hypersensitivity (DTH) response involves an initial sensitization by dermal administration of antigen (vaccination), followed a number of days later by the readministration of the antigen to the derma. Although the response can be simply measured as the degree of swelling, the influx of immune cells and cytokine production can also be determined. Dhabhar has investigated acute stress-induced immunoenhancement using the DTH response model with acute restraint stress in rodents. They have reported that restraint stress, either at the time of initial vaccination or at reexposure to antigen, enhanced the DTH response, measured by increased pinna thickness (site of antigen administration) in mice (for review, see (7)). Others have replicated and extended these findings. For example, a recent study confirmed the previously observed enhancement of the DTH response by acute restraint stress and, moreover, indicated a role for dendritic cells (26).
In summary, there is now consistent evidence from animal studies that both antibody and cell-mediated responses to vaccination are enhanced by acute stress at the time of antigen administration. If replicated in human studies, such effects could hold wide-ranging implications for vaccination strategies and public health immunization programs.
Early Human Studies
The relevant human evidence to support enhancement of the vaccine response to acute stress is sparse. A review of the literature identified three previous studies that have examined the effects of acute exercise or psychological stress on the response to vaccination in humans. Two were explicitly designed to assess the "open window" hypothesis, which predicts a period of immunosuppression, after a bout of high-intensity and long-duration exercise (12). Nevertheless, Eskola et al. (12) observed that the antibody response to tetanus toxoid was higher when four runners were vaccinated after a marathon race, compared with a nonrunner control group (4). This preliminary finding is contrary to the hypothesized impaired response after exercise after prolonged exercise. However, the second and considerably larger study found no differences in the antibody responses to tetanus toxoid, diphtheria, and pneumococcal antigens between 22 triathletes who were vaccinated after completing a half-ironman (3-km swim, 130-km cycle, and 21-km run) and 11 nonexercising triathletes and 22 sedentary controls (28). It is worth noting that both these previous exercise studies tested only men. The third study investigated the effects of psychological distress on the response to a novel antigen in young healthy students. Participants were vaccinated with KLH either immediately after a viva voce examination or during an examination-free period. Distress at time of vaccination was measured using the profile of mood states (POMS) questionnaire. The antibody response to KLH was measured 3-wk postvaccination. No difference in antibody response was observed between experimental groups nor was there a group difference in number of participants who developed an antibody response (defined as an enzyme-linked immunosorbent assay optical density of greater than three SD above the mean of six unvaccinated control subjects). The POMS scores differed significantly between groups. At vaccination, the group vaccinated after the examination reported more negative mood than the group vaccinated during an examination-free period; at follow-up, 3 wk later, however, it was the latter group that showed the greater negative mood. The POMS scores were not predictive of the antibody response to KLH (28). The paucity and inconsistency of studies that have addressed the influence of acute stress on the immune response to vaccination in humans, along with the compelling results from research on rodents, suggest that further dedicated studies are warranted.
RECENT HUMAN STUDIES
Our group has now conducted and published the results of two studies that used a randomized control design to investigate the immune effects of acute exercise and psychological stress before vaccination in humans. Two strategies were used to begin to elucidate the mechanisms by which acute stress may induce immunoenhancement. First, both studies used measures of the acute response to exercise as indicators of changes in the inflammatory milieu. Second, one study also included both a T-dependent (influenza) and a T-independent (meningococcal A + C) vaccine to determine the importance of T cell influence in any modulation.
Concentric Exercise and Mental Stress
The first study compared the effects of 45 min of either concentric cycling exercise, mental stress, and quiet reading before receiving the trivalent influenza vaccine and the meningococcal A + C vaccine (separately in contralateral arms) (9). The study also tested potential predictors of the antibody responses to try to identify possible mechanisms that might underlie any immunoenhancing effects of acute stress. Dhabhar implicated glucocorticoids in the enhancement of DTH. First, they showed that stress-induced enhancement of DTH responses was abolished by adrenalectomy; whereas, it remained in sham adrenalectomized and intact animals. Second, they found that administration of exogenous corticosterone at levels comparable to those found during acute stress, thus mimicking the effects of acute stress exposure, augmented the DTH response compared with control animals. It is possible, therefore, that elevated glucocorticoids at the time of antigen exposure play a role in the enhancement of the response.
Interleukin (IL) 6 is a key inflammatory responsive cytokine and another candidate for a potential role in acute stress-induced immunoenhancement. Interleukin 6 is one of the first cytokines to be elevated after exercise and is thought to help regulate the subsequent immune response (13). Interleukin 6 has been implicated in the vaccination response in previous rodent and human studies. Mice that were coadministered with the IL-6 gene with a DNA-based influenza vaccine were found to be completely protected from a challenge with a usually lethal dose of virus (17). In humans, when healthy adults were vaccinated with a live virus strain of Francisella tularensis, it was found that those classified as antibody responders had higher levels of IL-6 preimmunization than nonresponders (16). Thus, it was expected that any enhancement of antibody response by acute stress would be related to stress-induced increases in glucocorticoids and IL-6 at the time of vaccination.
In our study, the exercise group completed 45 min of moderate intensity cycling exercise. This level of exercise was chosen in part to avoid any possible immunosuppressive effects that are associated with high-intensity endurance exercise. In the "open window" hypothesis, long duration high-intensity exercise, such as marathons, have been hypothesized to cause a delayed immunosuppression that renders an athlete more susceptible to naturalistic infection (21). However, the threshold for these effects would not have been reached by the exercise completed in the current study.
In partial support of our hypothesis, the results confirmed an enhancement of the antibody response to one of the three influenza viral strains (A/Panama) in the exercise and psychological stress groups compared with control group (Fig. 1). However, this effect was only manifest in women; men showed no beneficial immune effect of exercise or mental stress. Sex differences in vaccination responses have been reported previously, and specifically, the A/Panama influenza strain has been found to generate stronger responses in men than in women (22). It is worth noting that women in the control group had shown poorer antibody responses compared with control men and, moreover, that exercise and mental stress had enhanced their otherwise poor responses. The A/Panama was the only strain to show any effect of acute stress; this strain was the least immunogenic, eliciting the smallest responses. Antigens that elicit a less robust response produce greater individual variation and provide the opportunity for small- to medium-sized effects to emerge (6), and thus, effects were most likely to be found in this strain.
Interestingly, men and women also showed different task-induced IL-6 response profiles that were consistent with our previous studies characterizing the cytokine response to exercise and psychological stress (10,11). Importantly, in hierarchical linear regression analysis, which controlled for baseline titers, the IL-6 level at 60 min after the end of the stress exposures was a significant predictor of A/Panama antibody titers 4 wk after vaccination in women. However, IL-6 was the only cytokine measured, and it is likely that is indicative of the general inflammatory response; as such, it is hypothesized that it is probably the overall cytokine milieu that is key, rather than IL-6 per se. The data provided no support for a role for glucocorticoids in acute stress-induced immunoenhancement. There are also inconsistent results for the role of cortisol in the context of the association between chronic stress and vaccination response (22,29), which emphasize that the role of cortisol in human immunity is complex and yet to be fully understood.
This first study also included two vaccines to allow the comparison of a T-dependent (influenza) and a T-independent (meningococcal A + C) vaccine response. Meningococcal polysaccharides elicit a thymus-independent antibody response in which activated B cells generate an antibody response without T cell help. In contrast to the influenza findings, in response to the meningococcal A serogroup, it was found that antibody responses were enhanced in men but not in women (Fig. 2). Therefore, the mechanism by which acute stress enhanced the antibody response to this polysaccharide vaccine must be independent of T cells. Although not yet investigated, this finding implies that the mechanism of the acute stress-induced immunoenhancement may lie at the interaction between B cells and native polysaccharide antigen, or through activation of complement on native antigen, or generation of the early antibody response. The common feature where enhancement of the immune response was found was the poor response of the control groups, evidenced in women in response to A/Panama and in men in response to meningococcal A. Thus, acute stress-induced immunoenhancement seems to be able to effectively boost a normally poor response.
The second study explored further the possibility that acute exercise exposure can act as an endogenous vaccine adjuvant. An eccentric exercise task was designed to initiate a local inflammatory response in the biceps brachii and deltoid muscles of the arm in which the influenza vaccine was administered (8). The antibody response was again assessed at 6 and 20 wk, but in addition, the cell-mediated response was also measured at 8 wk. T lymphocytes sensitized during initial antigen exposure are hypersensitive to the antigen on reexposure; an in vitro stimulation of whole blood with the vaccine causes a cellular cytokine (interferon (IFN) γ) response, which reflects the memory T cell response. Sex differences were again observed; exercise was associated with enhanced antibody responses in women but not in men (Fig. 3). The cell-mediated immune response tended to be higher overall in the exercise group than in the control group; however, this difference was driven by men who showed significantly greater cell-mediated responses after eccentric exercise compared with those in the control group (Fig. 4).
Eccentric exercise caused expected increases in arm circumference, which most likely reflects a combination of increased blood perfusion to the muscle, shortening of the myofibers, and inflammation-associated edema (24). There are several mechanisms by which such changes may influence subsequent immune responses. For example, increased tissue perfusion may increase lymphatic flow to draining lymph nodes, leading to a more efficient immune response. The local release of inflammatory mediators by damaged tissue cells, such as heat shock proteins, are known to activate dendritic cells, which is likely to facilitate their migration to the lymph nodes and enhance antigen-presenting capabilities (25). The percentage change in arm circumference was a strong predictor of the IFN-γ response in men in the exercise group, suggesting that the cell-mediated immune response might have been potentiated by increased local inflammation. In addition, the finding that men experienced greater percentage change in arm circumference than women indicated that men had a greater metabolic and/or immunologic response to exercise, and moreover, the latter may mediate the differential immune response to vaccination observed.
It is notable that the enhancement of responses to vaccines seems to be associated with poorer control responses rather than being a sex-specific effect; however, differences between men and women have been consistently shown in recent studies. Sex differences in the immune system are well accepted but not well understood. Generally, it is found that women mount more vigorous responses to infection or vaccination than men (5), but several studies have now reported greater responses in men than in women after diphtheria toxoid (1), measles (18), and influenza (22) vaccination. The mechanism of the sex difference is beyond the scope of the current data. However, it is known that some immune cells have estrogen and androgen receptors (2,23) and, importantly, that there are sex differences in the endocrine response to acute stress (15). Indeed, the inflammatory response to acute stress also exhibits sex differences (9-11). It is possible that these differences drive the helper T cell 1/helperT cell 2 (TH1/TH2) balance differently to influence the magnitude of the vaccination response. However, it should be recognized that the network of endocrine and neuronal factors that influence the TH1/TH2 balance is highly complex. Thus, speculation must be limited until the roles of stress hormones, catecholamines, and sex hormones are investigated further in acute stress-induced immunoenhancement in humans.
CLINICAL IMPLICATIONS AND CONCLUSIONS
The success of vaccines as a medical intervention against infectious disease illustrates their importance in preventative medicine. Since their inception, however, researchers have attempted to improve the efficacy of suboptimal vaccines and their capacity to protect against disease. Increasing the efficacy of a vaccine would allow optimal responses at lower doses, which would decrease costs and increase availability - an issue not only critical for developing countries but also worldwide in the case of possible pandemics. A common method of maximizing vaccine efficacy involves the use of adjuvants. Adjuvants act to increase the efficacy of the vaccine response by stimulating the innate immune system, which provides for a rapid response of the first line of defense against infection. Among the many effects of the innate response are a rapid burst of inflammatory cytokines and activation of antigen-presenting cells, which prepare the immune system for subsequent development of specific adaptive immune response to the vaccine. These characteristics of the activation of the innate immune system bear great similarity to the response elicited by acute stress, and as such, the enhancing effects summarized in this review have been proposed to act collectively as an endogenous adjuvant and are presented in the schematic in Figure 5. The data summarized here include the effects of both concentric and eccentric exercise and mental stress, but it is yet to be established what is the most effective type or duration of acute stress for enhancement of responses to vaccines. Indeed, further research is also needed in the populations who may benefit most from enhancement of vaccine responses, such as the elderly who show generally weaker responses. In summary, acute stress may enhance the response to vaccination in humans and has the potential to be developed and refined for use as a behavioral adjuvant.
1. Atabani, S., G. Landucci, M.W. Steward, H. Whittle, J.G. Tilles, and D.N. Forthal. Sex-associated differences in the antibody-dependent cellular cytotoxicity antibody response to measles vaccines. Clin. Diagn. Lab. Immunol.
2. Bebo, B.F. Jr, J.C. Schuster, A.A. Vandenbark, and H. Offner. Androgens alter the cytokine profile and reduce encephalitogenicity of myelin-reactive T cells. J. Immunol.
3. Binder, R.J., M.L. Harris, A. Menoret, and P.K. Srivastava. Saturation, competition, and specificity in interaction of heat shock proteins (hsp) gp96, hsp90, and hsp70 with CD11b+ cells. J. Immunol.
4. Bruunsgaard, H., A. Hartkopp, T. Mohr, H. Konradsen, I. Heron, C.H. Mordhorst, and B.K. Pedersen. In vivo
cell-mediated immunity and vaccination response following prolonged, intense exercise. Med. Sci. Sports Exerc.
5. Cannon, J.G., and B.A. St Pierre. Gender differences in host defense mechanisms. J. Psychiatr. Res.
6. Cohen, S., G.E. Miller, and B.S. Rabin. Psychological stress and antibody response to immunization: a critical review of the human literature. Psychosom. Med.
7. Dhabhar, F.S. Stress-induced augmentation of immune function-the role of stress hormones, leukocyte trafficking, and cytokines. Brain Behav. Immun.
8. Edwards, K.M., V.E. Burns, L.M. Allen, J.S. McPhee, J.A. Bosch, D. Carroll, M. Drayson, and C. Ring. Eccentric exercise as an adjuvant
to influenza vaccination in humans. Brain Behav. Immun.
9. Edwards, K.M., V.E. Burns, T. Reynolds, D. Carroll, M. Drayson, and C. Ring. Acute stress exposure prior to influenza vaccination enhances antibody response in women. Brain Behav. Immun.
10. Edwards, K.M., V.E. Burns, C. Ring, and D. Carroll. Individual differences in the interleukin-6 response to maximal and submaximal exercise tasks. J. Sports Sci.
11. Edwards, K.M., V.E. Burns, C. Ring, and D. Carroll. Sex differences in the interleukin-6 response to acute psychological stress. Biol. Psychol.
12. Eskola, J., 0. Ruuskanen, E. Soppi, M.K. Viljanen, M. Jarvinen, and H. Toivonen. Effect of sport stress on lymphocyte transformation and antibody formation. Clin. Exp. Immunol.
13. Febbraio, M.A., and B.K. Pedersen. Muscle-derived interleukin-6: mechanisms for activation and possible biological roles. FASEB J.
14. Gallucci, S., M. Lolkema, and P. Matzinger. Natural adjuvants: endogenous activators of dendritic cells. Nat. Med.
15. Kovacs, E.J., K.A. Messingham, and M.S. Gregory. Estrogen regulation of immune responses after injury. Mol. Cell Endocrinol.
16. Krakauer, T. Levels of interleukin 6 and tumor necrosis factor in serum from humans vaccinated with live, attenuated Francisella tularensis
. Clin. Diagn. Lab. Immunol.
17. Lee, S.W., J.W. Youn, B.L. Seong, and Y.C. Sung. IL-6 induces long-term protective immunity against a lethal challenge of influenza virus. Vaccine
18. Mark, A., R.M. Carlsson, and M. Granstrom. Subcutaneous versus intramuscular injection for booster DT vaccination of adolescents. Vaccine
19. Matzinger, P. The danger model: a renewed sense of self. Science
20. Millan, S., M.I. Gonzalez-Quijano, M. Giordano, L. Soto, A.I. Martin, and A. Lopez-Calderon. Short and long restraint differentially affect humoral and cellular immune functions. Life Sci.
21. Nieman, D.C., and B.K. Pedersen. Exercise and immune function. Recent developments. Sports Med.
22. Phillips, A.C., V.E. Burns, D. Carroll, C. Ring, and M. Drayson. The association between life events, social support, and antibody status following thymus-dependent and thymus-independent vaccinations in healthy young adults. Brain Behav. Immun.
23. Polan, M.L., J. Loukides, P. Nelson, S. Carding, M. Diamond, A. Walsh, and K. Bottomly. Progesterone and estradiol modulate interleukin-1 beta messenger ribonucleic acid levels in cultured human peripheral monocytes. J. Clin. Endocrinol. Metab.
24. Proske, U., and D.L. Morgan. Muscle damage from eccentric exercise: mechanism, mechanical signs, adaptation and clinical applications. J. Physiol.
25. Rock, K.L., A. Hearn, C.J. Chen, and Y. Shi. Natural endogenous adjuvants. Springer Semin. Immunopathol.
26. Saint-Mezard, P., C. Chavagnac, S. Bosset, M. Ionescu, E. Peyron, D. Kaiserlian, J.F. Nicolas, and F. Berard. Psychological stress exerts an adjuvant
effect on skin dendritic cell functions in vivo
. J. Immunol.
27. Silberman, D.M., M.R. Wald, and A.M. Genaro. Acute and chronic stress exert opposing effects on antibody responses associated with changes in stress hormone regulation of T-lymphocyte reactivity. J.Neuroimmunol.
28. Smith, A., U. Vollmer-Conna, B. Bennett, D. Wakefield, I. Hickie, and A. Lloyd. The relationship between distress and the development of a primary immune response to a novel antigen. Brain Behav. Immun.
29. Vedhara, K., N.K. Cox, G.K. Wilcock, P. Perks, M. Hunt, S. Anderson, S.L. Lightman, and N.M. Shanks. Chronic stress in elderly carers of dementia patients and antibody response to influenza vaccination. Lancet
30. Wood, P.G., M.H. Karol, A.W. Kusnecov, and B.S. Rabin. Enhancement of antigen-specific humoral and cell-mediated immunity by electric footshock stress in rats. Brain Behav. Immun.