| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ORIGINAL ARTICLES |
From the Department of Psychiatry (K.M.E.), University of California, San Diego, San Diego, California; School of Sport and Exercise Sciences (V.E.B., D.C., C.R.), University of Birmingham, UK; and the Department of Clinical Immunology (A.E.A., M.D.), School of Medicine, University of Birmingham, UK.
Address correspondence and reprint requests to Kate Edwards, School of Sport and Exercise Sciences, University of Birmingham, Edgbaston, B15 2TT, UK. E-mail: k.edwards.1{at}bham.ac.uk
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
Methods: Sixty healthy young adults were randomized to exercise (n = 20), mental stress (n = 20) or control (n = 20) before meningococcal A+C vaccination. Antibody concentration was measured by microsphere-based antibody quantification assay at prevaccination, 4 and 20 weeks post vaccination.
Results: Meningococcal serogroup A antibody responses were enhanced by exercise and mental stress in men but not women (F(2,51) = 4.00, p = .02, 
2 = 0.135).
Conclusions: Stress-induced immune enhancement has now been demonstrated in the antibody response to thymus-independent as well as thymus-dependent vaccines. These findings indicate that this effect is not specific to T-cell involvement.
Key Words: acute stress • exercise • adjuvants • meningococcal vaccination
Abbreviations: IgG = immunoglobulin G.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
Recently, in the first of our vaccine studies, we reported that 45 minutes of exercise stress (concentric cycling) or mental stress (time-pressured mental arithmetic with social evaluation) before vaccination enhanced women’s antibody response to the A/Panama strain of the influenza vaccine (6). The results also indicated that the enhanced response was positively associated with the interleukin-6 response to acute stress. Our subsequent study found that eccentric exercise (lowering weights) also enhanced the antibody response to influenza vaccination in women, whereas conversely this muscle-damaging exercise enhanced the cell-mediated immune response in men (7). It is worth noting that, in both studies, the stressors improved the immune response when the response of control subjects was relatively poor.
The purpose of the present investigation was to examine the effects of cycling exercise or mental stress on the antibody responses to the meningococcal A+C vaccine. This report is a follow-up to our earlier report that examined the effects of acute stress on the influenza vaccine (6); in the study, participants were administered consecutively the influenza vaccine and the meningococcal A+C vaccine into contralateral arms. The current report was made possible by our development of assays based on the microsphere-based meningococcal antibody quantification protocol of Lal and colleagues (8). The two vaccines were chosen to allow comparison of the sensitivity of thymus-independent (meningococcal A+C) and thymus-dependent (influenza) antigens to acute stress-induced immunomodulation. As such, inoculation with these two types of vaccines permitted assessment of the involvement of T cells in the effects of acute stress on antibody response. It was hypothesized that cycling exercise and mental stress would enhance antibody responses to the meningococcal A+C vaccine compared with resting control.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
Participants completed an initial testing session and 4-week and 20-week follow-up sessions. They were randomly allocated to one of three testing conditions (exercise stress, mental stress, or control) according to CONSORT guidelines. Twenty participants completed each condition and were tested in groups of five. They all attended the laboratory at 3 PM. After instruction, cannulation, and instrumentation, participants sat resting for 20 minutes (baseline) and completed questionnaires. Questionnaires assessed stressful life events in the previous year and previous month, psychological well-being and health behaviors (6). After the baseline, participants completed their assigned condition (45 minutes), after which a nurse administered consecutively meningococcal A+C (Mengivac A+C, Aventis Pasteur; Batch# W0483-4) and influenza vaccinations via intramuscular injections into the participants’ contralateral upper arms. They then rested for 60 minutes (recovery). At 4 and 20 weeks post vaccination, participants gave a single venous blood sample.
Tasks
Exercise Stress Task
Participants began by performing a four-step incremental cycle ergometer test, with each step lasting 4 minutes. The workloads performed at each step were 84, 133, 182, and 231 W for men and 70, 98, 126, and 154 W for women. These workloads were selected after pilot testing so as to elicit similar heart rates in men and women. This incremental component allowed the calculation of predicted maximum workload, based on their age-predicted maximum heart rate (9). The participants then performed an active recovery, in which they cycled for 4 minutes at 130 W (men) or 95 W (women), followed by 25 minutes at 55% predicted maximum workload. Participants were instructed to pedal at 70 revolutions per minute throughout. The cycle ergometers were located in a single large laboratory, allowing participants to see each other, but they were unaware of the workload of other participants and they were discouraged from communicating. This protocol allows single session testing without the need for prior fitness testing, at the same time standardizing the intensity to individual peak power output.
Mental Stress Task
The mental arithmetic task was an adapted version of the paced auditory serial addition task (PASAT) (10) modified to allow group testing. The PASAT has been shown to reliably perturb both cardiovascular and immune function (11–13). Participants were required to add two sequentially presented single-digit numbers and say the answer out loud. They retained the latter of the two numbers in memory for subsequent addition to the next number presented. Numbers, which ranged from 1 to 9, were delivered using an audio tape player. An experimenter sat 1 m in front of each of the five participants and checked each response against the correct answers. Participants were able to see themselves in a mirror placed behind the experimenter’s right shoulder and were instructed to look at themselves in the mirror at all times; this was designed to heighten social anxiety. Participants received a brief burst of loud, aversive noise every ten trials. They were informed this "punishment" was performance related. In each block of ten numbers, participants received a noise burst after only their first incorrect response. However, if they had not made an error, they received a burst at the end of the block. This protocol ensures that the same exposure to aversive stimuli is experienced by all participants. By testing five people at once, elements of distraction, competition, and social evaluation were introduced. Participants sat in a row 1 m apart and could hear the simultaneous answers of others and the aversive noises to which the other participants were subject. The five participants competed in the 45-minute task, which consisted of four 8-minute contests with a 3-minute rest between each contest when the scores were calculated. After each contest, the placings were announced and points were awarded accordingly; this information was then displayed on a leader board. Each contest became progressively harder as the speed of presentation of numbers increased through each contest and from one contest to the next. The first contest consisted of four consecutive 2-minute periods of 33, 38, 43, and 50 digits, respectively, at presentation rates of 3.6, 3.2, 2.8, and 2.4 seconds. The presentation rates during the initial 2-minute period of the second, third, and fourth contests were 3.2, 2.8, and 2.4 seconds, respectively, and, as in the first task, the presentation rates increased by 0.4 second every 2 minutes. In the same way, the points available for each contest increased such that the final contest was always worth sufficient points for overall victory. Cash prizes were awarded to the participants in 1st (£5), 2nd (£3) and 3rd (£2) places overall.
Control Task
Participants sat and read quietly for 45 minutes.
Blood Sampling
Blood samples were collected from participants via an intravenous catheter (18-gauge, Teflon, Insyte, Becton Dickinson, UK) in an anticubital vein during the testing session, and by venepuncture at follow-up sessions. At the testing session, after the resting baseline, and at both follow-up sessions, blood samples were drawn into two 7 ml plain tubes, allowed to clot for 1 hour, centrifuged (3500 rpm for 5 minutes), and the serum was aliquoted and stored at –20°C for antibody analysis.
Meningococcal A and C Antigen Specific Immunoglobulin G (IgG) Assays
A microsphere-based multiplexed assay for quantification of serum antibodies against meningococcal serogroups A and C was developed in our laboratory, based on the protocol described by Lal et al. (8) with minor changes detailed here. Neisseira meningitidis capsular polysaccharides for serogroups A and C were attached to poly-L-lysine and conjugated to activated carboxyl-modified microspheres in a two-step carbodiimide reaction. Samples and standards were all tested in triplicate. The standard reference serum sample CDC1992 (NIBSC, UK) was used to create a standard curve. Concentrations of meningococcal A and C antibodies in CDC1992 were used in this assay (14,15). Microspheres were used at a prediluted concentration of 5000 beads/region/well, and the secondary antibody used was R-phycoerythrin-conjugated antihuman IgG (Southern Biotech, UK). Specificity and recovery tests produced responses >80% and 90%, respectively. Intra-assay variation was low for both meningococcal A and C at 8.5% and 7.4%, respectively.
Statistical Analysis
The distributions of the antibody concentrations for both meningococcal A and C at all three time points were significantly different from normal, according to both Kolmogorov-Smirnov and Shapiro-Wilk tests of normality (p < .002). Due to these skewed distributions, analyses were performed on log10 transformed values. Although the focus was on the meningococcal A response, the vaccine also included meningococcal C antigens; therefore, the effects of stress on the response to both strains were analyzed. However, some participants (n = 14) reported that they had never previously received the meningococcal C conjugate vaccine. Primary antibody responses are smaller than secondary antibody responses. A 2 Group (previously vaccinated, unvaccinated) x 3 Period (baseline, 4 weeks, 20 weeks) repeated- measures multivariate analysis of variance (MANOVA) revealed a significant interaction (F(2,54) = 4.50, p = .02, 2 = 0.143), which confirmed that unvaccinated individuals showed significantly lower meningitis C antibody responses than previously vaccinated participants, with levels at both 4 and 20 weeks lower than previously vaccinated individuals. Thus, the unvaccinated participants were excluded from the analysis of meningococcal C antibody data.
The baseline carrier rates seen in the current study were similar for both meningococcal A and C. Although we controlled for prior vaccination, it is not possible to control for naturalistic exposure. It is known that carrier rates for meningococci are likely to be >10% in Europe (16), and naturalistic exposure to bacteria other than N. meningitides group A can cause an increase in antibodies specific to meningococcal serogroup A (17). However, in confirmation of the primary nature of the meningococcal A response and the secondary nature of the meningococcal C response, antibody responses were larger to serogroup C than to serogroup A. This is in line with findings that prior vaccination with conjugated serogroup C primes for robust memory responses, whereas exposure to polysaccharide alone does not (18,19).
The analytic strategy followed that described by Edwards et al. (6) for the influenza responses. A series of 2 Sex x 3 Condition (exercise stress, mental stress, control) x 3 Period (baseline, 4 weeks, 20 weeks) repeated-measures MANOVAs were performed on the antibody data. The focus was the Sex x Condition x Period interaction effect; planned orthogonal polynomial contrasts were performed on these interactions given that we were concerned with variations among conditions in the patterning of antibody data across the three time periods. Planned comparisons, co-varying for baseline antibody levels, were employed to determine differences between groups. Eta-squared (2), a measure of effect size, was determined. Occasional missing data are reflected in the reported degrees of freedom.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
2 = 0.135). The pattern showed that men in exercise and mental stress groups had larger increases in IgG antibodies than in the control group, whereas women showed similar responses in all groups. As sex differences in antibody responses have been previously shown and a significant overall interaction was found, MANOVA was preformed separately for men and women. A significant Condition x Period interaction was found for men (F(4,52) = 2.65, p = .04, 2 = 0.169), but not for women (F(4,50) = 1.15, p = .35, 2 = 0.084). Planned comparisons of antibody levels at 4 weeks post vaccination, co-varying for baseline, found a significant effect of condition for men but not for women. At 20 weeks, no significant effects were found. For men, exercise and mental stress were associated with improved responses compared with control (Figure 1A). However, for women, who had a stronger response in the control condition, exercise or mental stress did not significantly alter the response.
|
Meningococcal C Antibody Levels
Analysis of the meningococcal serogroup C data showed the expected effect of period (F(2,36) = 43.70, p < .001, 2 = 0.702) with antibody levels increasing from baseline to 4-week follow-up and remaining higher than baseline at the 20-week follow-up. However, no significant Sex, Period, or Sex x Period interaction effects emerged. As participants who had never been previously immunized with the conjugate meningococcal C vaccine were excluded, the responses represent secondary responses in which participants in all conditions showed similarly robust antibody responses (Figure 1C and D).
Potential Confounding
Previous research has documented that the antibody response to vaccination has been associated with demographic, psychological, and behavioral factors (1–3). It is possible, therefore, that the immunoenhancement observed for meningococcal A in men could be accounted for by variations in these factors. However, there were no differences among conditions in age, body mass index, life event exposure, perceived stress, psychological morbidity, or health behaviors.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
One possibility is that stress-induced immunoenhancement may manifest when the control participants mount a poor immune response to the vaccination. On each occasion that there was a clear difference between the immune responses of men and women in the control groups, there was a stress-induced enhancement of the immune response that was confined to the sex who had mounted the poorer control response. For example, women exhibited stress-induced enhancement when the antibody response to influenza vaccination was otherwise poor (6,7), whereas men’s responses were improved by stress when their cell-mediated immune response to influenza vaccination (7) or antibody response to meningococcal A vaccination was otherwise poor. Further, no evidence of stress-induced immunoenhancement was found when there were no sex differences in the responses of the control groups. Taken together, these findings suggest that the immunoenhancing effects of acute stress are manifest when the standard control response is less robust. This is consistent with the work by Cohen et al. (2), who noted that antigens that elicit less robust responses provide greater individual variation and thus allow small- to medium-sized effects to emerge. This hypothesis must be tested using a vaccine that elicits a poor response in both sexes; we would speculate that, in this scenario, both male and female participants would demonstrate stress-induced immunoenhancement.
The finding that meningococcal A antibody responses were enhanced by acute stress provides further evidence relevant to understanding the mechanisms of such immunoenhancement. Meningococcal polysaccharides elicit a thymus-independent antibody response, in which activated B cells generate an antibody response without T-cell help (20,21). 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 effects may lie at the interaction between B cells and native polysaccharide antigen, or through activation of complement on native antigen, and/or other aspects of the innate immune response. Previously, we reported that indices of the stress response, interleukin-6 (6) and eccentric exercise-induced edema (7), were associated with enhanced responses to influenza vaccination. However, analyses found no associations between meningococcal antibody responses and either interleukin-6 or cortisol responses to acute stress in the current study (data not presented) (6). Further mechanistic studies, with more detailed measurements of cytokines, hormones, and T- and B-cell markers, are required to determine the pathways responsible for the effects of acute stress-induced immunoenhancement.
The findings of the current study should be interpreted in light of a number of potential limitations. First, the co-administration of the influenza vaccine and the meningococcal A+C vaccine could be considered a methodological limitation, due to the possibility of confounding between the two simultaneous immune responses. However, in clinical settings, thymus-dependent and thymus-independent vaccines are administered routinely in the same visit to the health practitioner. For example, when both the influenza vaccine and the polysaccharide pneumovax vaccine are required, co-administration into separate arms is recommended in current UK National Health Service protocol. Indeed, the safety (22), and in this instance, more importantly the efficacy (23,24) of simultaneous administration with influenza and pneumococcus vaccines has been shown to be unaltered compared to single administration. Further, this criticism can be refuted from a mechanistic perspective. In response to the influenza vaccine, T-helper cells specific to the strain of antigen supply cognate help to B cells specific to the same antigen. Thus, these cells could not interact with B cells specific for the meningococcal A or C serogroups. The only possible route for this sort of interference would be if the antigens became linked; for example, if both became linked to alum. In this case, a meningococcal A or C specific B cell might internalize an influenza antigen, and subsequently be able to recruit T-helper cells specific to the influenza antigen. However, as the vaccines were separately administered to contralateral arms, it is highly unlikely to have occurred. Regardless of these factors, these findings should be replicated with each vaccine administered in isolation to confirm that the administration of two vaccines did not cause an unwanted interaction.
A second potential limitation is that the current study does not include a detailed analysis of the kinetics of the antibody response to the vaccinations. However, the peak antibody response is well accepted to occur between 4 and 6 weeks post vaccination, and levels subsequently decline toward baseline. As such, the points measured in the current study are likely to reflect the peak response and the degree of maintenance. Future studies could incorporate more frequent follow-up assessments, to better capture the effects of the intervention on the immune response profile. Finally, it must be acknowledged that the current study used only young, healthy adults and therefore the generalizability of the findings to other populations remains to be determined. Further investigation of this phenomenon in more vulnerable populations is warranted, as effective interventions using behavioral adjuvants would be particularly beneficial for individuals who typically mount poor responses to vaccinations, such as the elderly.
In sum, the present data provide further evidence that acute stress at the time of antigen exposure can enhance the response to vaccination in humans. Although the data suggest that the adjuvant effects of stress are not limited to responses requiring T-cell help, further studies are required to help better understand the mechanisms underlying stress-induced immunoenhancement. It is also important to optimize the interventions to maximize their efficacy and feasibility for large-scale implementation. However, the current data add to the accumulating evidence that behavioral adjuvants, such as acute stress around the time of vaccination, have the potential to help improve the clinical efficacy of vaccinations.
We thank Dr. Mark Cobbold and Dr. Jos Bosch for valuable advice during assay development, and Dr. Jet Veldhuijzen van Zanten, Dr. Anna Phillips, Ms. Tracy Reynolds, and Ms. Abigail Kay for help during data collection.
|
NOTES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
Received for publication February 15, 2007; revision received September 27, 2007.
DOI:10.1097/PSY.0b013e318164232e
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  F. S. Dhabhar A hassle a day may keep the pathogens away: The fight-or-flight stress response and the augmentation of immune function Integr. Comp. Biol., September 1, 2009; 49(3): 215 - 236. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
