|Home | About | Journals | Submit | Contact Us | Français|
Evidence suggests that optimism may be protective for health during times of heightened stress, yet the mechanisms involved remain unclear. In a double-blind placebo-controlled study, we recently showed that acute psychological stress and an immune stimulus (Typhim-Vi typhoid vaccine) synergistically increased serum levels of interleukin-6 (IL-6) and negative mood in 59 healthy men. Here we carried out further analysis of this sample to investigate the relationship between dispositional optimism and stress-induced changes in immunity and mood. Volunteers were randomly assigned to one of four experimental conditions in which they received either typhoid vaccine or saline placebo, and then rested or completed two mental tasks. In the stress condition, optimism was inversely related to IL-6 responses, independent of age, BMI, trait CES-D depression and baseline IL-6. This relationship was present across both stress groups (combining vaccine and placebo) and was not present in the vaccine/stress group alone, suggesting that optimism protects against the inflammatory effects of stress rather than vaccine per se. Typhoid vaccine induced a significant increase in participants’ circulating anti-Vi antibody levels. Stress had no effect on antibody responses overall. However, in the vaccine/stress group, there was a strong positive association between optimism and antibody responses, indicating that stress accentuated the antibody response to vaccine in optimists. Across the complete sample, more optimistic individuals had smaller increases in negative mood and less reduction in mental vigour. Together these findings suggest that optimism may promote health, by counteracting stress-induced increases in inflammation and boosting the adjuvant effects of acute stress.
Dispositional optimism, defined as the generalized expectation that good rather than bad things will occur in one’s life, has been related to better psychological and physical well-being, particularly during times of heightened stress (Scheier and Carver, 1992; Smith and MacKenzie, 2006). Optimistic individuals recover more quickly following cardiac-related events such as coronary artery bypass surgery and myocardial infarction, with a more rapid return to a normal lifestyle and a better reported quality of life (Agarwal et al., 1995; Scheier et al., 1989). Optimism also appears to be associated with lower levels of distress, slower disease progression and improved survival rates in patients with HIV (Ironson and Hayward, 2008), and certain types of cancer (Allison et al., 2003; Carver et al., 2005; de Moor et al., 2006), and comparable protective effects of optimism on psychological and physical well-being have been reported in caregivers (Hulbert and Morrison, 2006), undergraduate students adapting to the first college semester (Segerstrom et al., 1998), pregnant women (Fontaine and Jones, 1997) and elderly people (Giltay et al., 2006a). In line with these findings, optimists have a significantly reduced risk of future cardiovascular as well as all-cause mortality (Chida and Steptoe, 2008; Giltay et al., 2006b; Kubzansky et al., 2001), an effect that is independent of traditional cardiovascular risk and socio-demographic factors.
The mechanisms underlying the potential protective effect of optimism on health remain unclear. Optimistic people tend to engage in more positive health practices, with higher levels of physical activity, reduced levels of smoking and alcohol consumption, and a healthier diet (Scheier and Carver, 1992). They are also more likely to adopt an active coping style when faced with adversity, and less likely to engage in avoidant coping, a strategy related to poorer long-term health (Taylor and Stanton, 2007). Furthermore, optimism is inversely correlated with negative personality characteristics such as neuroticism, anxiety and depression that are themselves related to poorer mental and physical health. However, despite their recognised contribution to the protective effects of optimism, the relationship between optimism and health persists after controlling for these factors, indicating that other pathways are involved.
One such pathway could be the immune system. Chronic stress is associated with dysregulated immunity and an increased susceptibility to cancer, autoimmune and inflammatory disease (Kemeny and Schedlowski, 2007), and there is emerging evidence that optimism can moderate the negative impact of stress on immunity. For example, during situations of academic stress that are commonly accompanied by reduced lymphocyte levels and function, students scoring high on optimism display greater numbers of cytotoxic T cells, improved natural killer cell (NKC) activity and larger antigen-stimulated delayed-type hypersensitivity responses than students with low optimism scores (Segerstrom et al., 1998; Segerstrom, 2001, 2005). Similarly, in women recently diagnosed and surgically treated for breast cancer, high levels of dispositional optimism were found to counteract the association between perceived stress and lowered NKC activity (Von Ah et al., 2007). However, the relationship between optimism and immunity is complex and dependent on the duration and type of stressor involved, and not all studies report a protective effect of optimism on stress-induced immune changes (Cohen et al., 1999a; Segerstrom, 2001, 2005). Furthermore, little is known about the potential effect of optimism on other stress-sensitive immune molecules such as cytokines and antibodies.
The inflammatory cytokine, interleukin-6 (IL-6) plays a key role in infectious and inflammatory disease (Naugler and Karin, 2008). Circulating levels of IL-6 are increased during immune activation, and correlate with infection-related somatic and depressive symptoms in humans and animals exposed to bacterial or viral pathogens (Cohen et al., 1999b; Dantzer et al., 2008; Reichenberg et al., 2001; Wright et al., 2005). Notably, psychological stress also increases circulating IL-6 (Steptoe et al., 2007), and acute psychosocial stressors were found to upregulate cytokine and behavioural responses to immune stimuli in rodents (Gandhi et al., 2007; Gibb et al., 2008). We recently carried out a double-blind, randomized, placebo-controlled study to examine whether acute psychological stress might have a similar effect on inflammatory cytokine and mood responses to an immune stimulus in humans (Brydon et al., 2009). In that study, healthy men were assigned to one of four experimental conditions in which they received typhoid vaccine or saline placebo and then either rested or completed two challenging behavioral tasks. Tasks induced significant increases in participants’ subjective stress ratings, systolic BP, diastolic BP and heart rate, with no differences between vaccine and placebo conditions. Typhoid vaccine but not placebo induced a significant increase in participants’ serum levels of IL-6, and this response was larger in the stress compared to rest conditions, indicating that stress accentuated the IL-6 response to vaccine. Total negative mood, assessed using the Profile of Mood States questionnaire, increased immediately following stress tasks, an effect also more pronounced in the vaccine/stress versus placebo/stress condition. Together, these findings suggested that, similar to observations in animals, psychological and immune stressors may act synergistically to promote inflammation and sickness behavior in humans (Brydon et al., 2009). In the current investigation, we carried out additional analyses of the same sample to examine the relationship between dispositional optimism and changes in IL-6 and negative mood following exposure to acute psychological stress and typhoid vaccine. We hypothesised that optimism would exert a protective effect by attenuating the synergistic and/or individual effects of psychological and immune stressors on IL-6 and negative mood in men. We also examined the potential association between optimism, psychological stress and antibody responses to the vaccine. Previous reports have shown that acute psychological stress accentuates antibody responses to other types of vaccines including influenza and Meningococcal A vaccine in humans (Edwards et al., 2006, 2008). We predicted that acute stress would enhance antibody responses to typhoid vaccine and that optimism would promote this effect.
Fifty-nine male volunteers were recruited for the study, the details of which are reported elsewhere (Brydon et al., 2009). All participants gave their informed consent and the study was approved by the joint UCL/UCLH Committee on the Ethics of Human Research. Participants were students at University College London, aged between 18 and 30 years, and were screened by structured interview to ensure that they were healthy, had no previous history of any relevant physical or psychiatric illness, were taking no medication and were non-smokers. They were instructed not to take antibiotics, aspirin or ibuprofen for 14 days prior to testing, and to avoid caffeinated beverages, alcohol and excessive exercise during the 12 h prior to testing. Volunteers who had received typhoid vaccine in the past 3 years or any other vaccine in the previous 6 months were excluded.
Salmonella typhi capsular polysaccharide vaccine (Typhim-Vi, Aventis Pasteur) was used as an immune stimulus. This vaccine is a thymus-independent type 2 antigen (Kroon et al., 1999) previously shown by our group and others to increase circulating levels of cytokines and induce negative mood states in healthy volunteers (Strike et al., 2004; Wright et al., 2005). Unlike other infection models, it is a relatively mild inflammatory stimulus and does not provoke fever or feelings of malaise that could potentially confound responses measured.
Dispositional optimism was assessed using the revised version of the Life Orientation Test (LOT-R), a 10-item self report questionnaire that evaluates generalised expectations of positive and negative outcomes (Scheier et al., 1994). Participants were asked to indicate the extent of their agreement with each item from 0 (strongly disagree) to 4 (strongly agree). Only six items are used to derive an optimism score (four are filler items). Thus, potential scores ranged from 0 to 24, with higher scores indicating higher levels of optimism. Mood and symptoms of illness were assessed with a modified 36-item version of the Profile of Mood States (POMS), as described previously (Brydon et al., 2009; McNair et al., 1981). Six high-loading items were taken from the vigour, tension–anxiety, depression–dejection, and confusion scales of the original POMS, and five items were taken from the fatigue scale. Four extra items were added to assess symptoms associated with mild infection (fever, aching joints, nausea, and headache). Participants were asked to rate how they felt at that moment on a 5-point scale from 0 = not at all to 4 = extremely. Total mood score was calculated (as recommended in the POMS manual) by summing all negative items (tension, depression, confusion and fatigue). Overall mood scores could range from 0 to 96, with higher scores indicating a more negative mood. Depression was assessed by the Centre for Epidemiological Studies Depression (CES-D) Scale, a widely used self-report questionnaire (Radloff, 1977). Scores could range from 0 to 60, with higher scores indicating more depressed mood.
Circulating concentrations of interleukin-6 (IL-6) were assessed in serum. Whole blood samples (10 ml) were drawn using a 21-gauge butterfly needle into a serum-separator vacutainer tube (BD Vacutainer Systems, Oxford, UK) and left upright to clot for 30 min. Samples were then centrifuged at 1250g for 10 min at room temperature and serum was removed, aliquoted and stored at −80 °C prior to analyses. Serum IL-6 concentrations were assessed in duplicate samples by a high sensitivity two-site enzyme-linked immunosorbent assay (ELISA) from R and D Systems (Oxford, UK). The detection limit of this assay was 0.09 pg/ml, with intra- and inter-assay coefficients of variation (CVs) of 4.69% and 4.66%, respectively.
Antibodies specific for S. typhi capsular polysaccharide vaccine (anti-Vi antibodies) were assessed in plasma. Whole blood samples (10 ml) were drawn using a 21-gauge butterfly needle into an EDTA-vacutainer tube (BD Vacutainer Systems, Oxford, UK), then centrifuged immediately at 1250g for 10 min at room temperature. Plasma was removed, aliquoted and stored at −80 °C prior to analyses. Anti-Vi antibodies were measured using an ELISA developed at the Laboratory of Enteric Pathogens, Health Protection Agency in London, UK as described previously (Chart et al., 2007).
Blood pressure and heart rate were assessed throughout the session using a Portapres-2, a portable version of the Finapres continuous BP monitoring device that shows good reproducibility and accuracy in a range of settings (TNO-TPD Biomedical Instrumentation, Amsterdam, Holland). Saliva samples were collected using Salivettes (Sarstedt Inc., Leicester, UK) and stored at −80 °C prior to analyses. Salivary cortisol was measured in duplicate samples by an ELISA (SLV-2930, DRG International, Inc., USA) at Kurume University in Japan. This assay had a detection limit of 0.53 ng/ml, and intra- and inter-assay CVs of 2.61% and 3.63%, respectively.
Details of this laboratory testing procedure have been reported previously (Brydon et al., 2009). The investigation was carried out in a double-blind, randomised, placebo-controlled manner. All sessions commenced at 09:00 and participants were assessed individually. Anthropometric measures were obtained using standardised methods and a baseline blood sample was drawn by separate venepuncture for assessment of basal circulating levels of IL-6 and anti-Vi antibodies. Participants were then randomly assigned to one of four experimental conditions (15 vaccine and stress; 15 placebo and stress; 14 vaccine and rest; 15 placebo and rest), by an investigator who had no involvement in participant testing. A trained nurse or clinician administered injections of either S. typhi capsular polysaccharide vaccine (0.025 mg in 0.5 ml, Typhim Vi, Aventis Pasteur, MD) or control saline placebo (0.5 ml) intramuscularly into the non-dominant deltoid muscle at approximately 09:45. Finger cuffs were fitted so that blood pressure and heart rate could be monitored using the Portapres-2, and participants were left to rest for 30 min. Blood pressure and heart rate were recorded for the last 5 min of the rest period, and a saliva sample was obtained for assessment of cortisol. Participants then either continued to rest or completed two challenging mental tasks, each lasting 5 min. They were not told whether they would rest or perform tasks until this time point. The first was a computerized Stroop task, involving the successive presentation of target color words printed in an incongruous color. Participants were instructed to press a key that corresponded to the position at the bottom of the screen of the name of the color in which the target word was printed. The second was a simulated public speaking exercise, in which participants were presented with an imaginary scenario where they had been wrongly accused of theft. They were instructed to give a speech in their defense and told that their speech would be video taped and subsequently assessed for efficacy and fluency. Five-minute recordings of blood pressure and heart rate were made during each of the tasks and a second cortisol sample was obtained immediately after the second task (approximately 10:45). Participants then rested quietly for the remainder of the session. They were asked to rate subjective feelings of stress on a 7-point scale from 1 = low to 7 = high at baseline (towards the end of the rest period), following each of the tasks and then at 30, 60 and 120 min post-task. Ratings of task difficulty, controllability and involvement were also made after each task on 7-point scales. Further recordings of blood pressure and heart rate were made at 25–30 min, 55–60 min and 115–120 min post-tasks and saliva samples were obtained at 15, 30, 60 and 120 min post-tasks. Subjective ratings of mood and illness symptoms were also obtained at baseline, immediately post-tasks, and at 30, 60 and 120 min post-tasks, and body temperature was measured at these time points using a sublingual digital thermometer. A second blood sample was drawn by separate venepuncture at 120 min post-tasks (corresponding to 3 h post-vaccination) for assessment of interleukin-6, and a third blood sample was obtained 3 weeks post-vaccination for assessment of anti-Vi antibodies.
Background characteristics of participants in the four groups were compared using analysis of variance for continuous variables, and χ2-tests for categorical variables. Subjective and cardiovascular responses to the experimental protocol were analysed using repeated measures analysis of variance, with group (vaccine/placebo and stress/rest) as between-person factors and trial (baseline, Stroop, Speech, recovery 1 (30 min), recovery 2 (60 min) and recovery 3 (120 min)) as the within-person factor. The repeated measures analysis of cortisol involved six trials (baseline, task, and 15, 30, 60 and 120 min post-task), while the analysis of serum IL-6 involved baseline and 120 min post-task measures, and analyses of plasma anti-Vi antibodies involved baseline and 3 week post-vaccination measures. The analyses of mood responses included vaccine and stress grouping factors and five trials (baseline, immediately post-task, and 30, 60 and 120 min post-task). The Greenhouse–Geisser correction of degrees of freedom was applied when sphericity assumptions were violated, but raw degrees of freedom are presented in Section 3. The distribution of serum IL-6 was skewed, so was log transformed prior to analysis. Post hoc tests were conducted using Tukey’s least significant difference (LSD) test. Associations between responses and optimism were analysed by multiple linear regression. The covariates in the regressions on mood responses included the baseline level of the mood variable, age and BMI, while scores on the CES-D were added to the regressions on biological responses to account for background affect. Effect sizes for optimism are represented by the change in r2 accounted for, over and above the effect of covariates.
Details of participant characteristics and general stress responses have been reported previously elsewhere (Brydon et al., 2009). There were no significant differences between the four experimental groups in terms of age, BMI, or baseline levels of cardiovascular activity, cortisol, immune measures or negative mood. Participants’ dispositional optimism scores on the LOT-R ranged from 2 to 22, with a mean score of 14.9, and there were no group differences in optimism scores. Scores on the CES-D averaged 12.59 ± 7.7, and also did not differ between groups. Participants’ body temperature remained stable throughout the session and none of the participants reported any appreciable somatic symptoms such as aching joints, headache or nausea.
Across all four groups, optimism was associated with a smaller increase in total negative mood at 30 min, controlling for baseline negative mood, BMI and age (β = −0.29, SE = 0.13, p = 0.030, additional r2 = 0.044). This effect was significant in the overall sample only and did not vary between groups. Analysis of scores on the POMS subscales showed that, in the complete sample, optimism was associated with a larger increase in vigour at 30 min, controlling for baseline vigour, BMI and age (β = 0.34, SE = 0.13, p = 0.011, additional r2 = 0.035) and with smaller increases in tension/anxiety at both 30 min (β = −0.25, SE = 0.11, p = 0.026, additional r2 = 0.075) and 60 min (β = −0.25, SE = 0.11, p = 0.032, additional r2 = 0.071), controlled for baseline tension/anxiety, BMI and age. The effect for tension/anxiety at 60 min is illustrated in Fig. 1 showing larger increases in tension/anxiety in people with lower optimism scores, adjusted for covariates.
Although stress alone had no overall effect on participants’ serum IL-6 levels, there were large individual differences in the IL-6 response, ranging from a decrease of 0.72 pg/ml to an increase of 4.01 pg/ml at 2 h post-task in the placebo/stress group. Similarly, IL-6 responses in the vaccine/stress group ranged from a decrease of 0.56 pg/ml to an increase of 3.10 pg/ml at 2 h. Optimism was inversely related to IL-6 stress responses (β = −0.50, SE = 0.24, p = 0.046, additional r2 = 0.158), independent of baseline IL-6 levels, BMI, age and trait CES-D depression. This effect is illustrated in Fig. 2 showing mean changes in serum IL-6 in relation to quintiles of optimism score, adjusted for covariates. Notably, this relationship was present in the stress groups (combining vaccine and placebo), and was not present in the vaccine/stress group alone, suggesting that optimism moderates the inflammatory response to psychological stress rather than vaccine per se. There was no relationship between optimism and IL-6 measures in either of the rest (vaccine/rest, placebo/rest) conditions (all p > 0.10).
Participants who had received typhoid vaccine but not placebo, had significant increases in plasma levels of anti-Vi antibodies 3 weeks following vaccination (F(1.54) = 57.7, p < 0.001) (Fig. 3). There was no significant difference in the stress and rest conditions in this response, suggesting that acute stress did not moderate the antibody response to typhoid vaccine. However, in the vaccine/stress group only, there was a strong positive association between optimism and antibody responses (β = 0.72, SE = 0.23, p = 0.013, additional r2 = 0.463), independent of baseline antibody level, age and BMI (Fig. 4). This effect was not present in any of the other 3 groups (β = −0.04 to +0.01, p > 0.40). Thus, in optimistic individuals, acute stress appears to accentuate antibody responses to typhoid vaccine. When trait depression (CES-D score) was included in the analysis, the relationship between optimism and antibody response was weakened (β = 0.58, p = 0.081), suggesting that this effect was partly dependent on the inverse correlation between optimism and depressed mood (r = −0.65). Nevertheless, there was no significant association between CES-D scores themselves and antibody responses (p = 0.42).
There were no significant associations between optimism and subjective ratings of stress or task difficulty, involvement, or perceived performance in either the vaccine/stress or placebo/stress groups, and no significant associations between optimism and measures of systolic BP or heart rate in any of the groups. However, in the vaccine/stress group, more optimistic individuals had significantly lower levels of diastolic BP at both 60 min (β = -0.80, SE = 0.20, p = 0.004, additional r2 = 0.383) and 120 min (β = −0.71, SE. = 0.22, p = 0.011, additional r2 = 0.300) post-stress, independent of baseline DBP, age, BMI and CES-D score. The effect at 120 min is illustrated in Fig. 5 (r = −0.70, p = 0.004). These associations were not significant in any other group (β = −0.102 to 0.116, all p > 0.60). In the vaccine conditions (vaccine/stress and vaccine/rest combined), there was also an inverse association between optimism and cortisol levels at 120 min post-stress (β = −0.45, SE = 0.20, p = 0.035, additional r2 = 0.097), independent of baseline cortisol levels, age, BMI and CES-D score. This effect is illustrated in Fig. 6, showing mean decreases in cortisol in relation to quintiles of dispositional optimism score, adjusted for covariates. There was no significant association between optimism and cortisol levels in the placebo groups (all p > 0.20).
One of the main findings of our study was that participants with high levels of dispositional optimism had smaller IL-6 responses in the stress condition. This relationship was observed across both stress groups (combining vaccine and placebo) but was not seen in the vaccine/stress group alone, indicating that optimism protects against the inflammatory effects of psychological stress rather than vaccine per se. Supporting this idea, a separate study investigating the relationship between optimism and stress responses in temporomandibular disorder (TMD) patients, found that optimistic patients had smaller plasma IL-6 levels during a speech stressor compared to their pessimistic counterparts (Costello et al., 2002). Other studies have also related positive psychological constructs to smaller stressor-induced increases in inflammatory markers. For example, positive affect was associated with decreased production of IL-6 by LPS-stimulated human monocytes in vitro (Prather et al., 2007) and smaller plasma fibrinogen responses to acute psychological stress in civil servants from the Whitehall II epidemiological cohort (Steptoe et al., 2005), and a recent investigation found an inverse relationship between self-esteem and inflammatory cytokine responses to acute psychological stress in women (O’Donnell et al., 2008).
Reduced stress-induced increases in IL-6 may be a mechanism through which optimism is protective for physical and mental health. IL-6 plays a pivotal role in inflammation and immunity, and circulating levels of this cytokine are elevated in a number of inflammatory-type conditions known to be exacerbated by stress including cardiovascular disease, arthritis, multiple sclerosis, asthma, pain and certain types of cancers (Kemeny and Schedlowski, 2007; Naugler and Karin, 2008). Although few studies have addressed the clinical relevance of acute increases in IL-6, IL-6 responses to a single bout of laboratory stress have been related to heightened pain sensitivity in TMD patients (Costello et al., 2002) as well as predicting future increases in blood pressure in healthy adults from the Whitehall 11 cohort (Brydon and Steptoe, 2005). Experimental infection studies have also demonstrated a key role for IL-6 in mediating illness symptoms. In adult volunteers exposed to influenza A virus, high daily levels of perceived stress were associated with elevated nasal concentrations of IL-6 and greater numbers of subjective and objective respiratory illness symptoms (Cohen et al., 1999b). In that study, the IL-6 response to infection was temporally related to both symptoms and mucus production, suggesting that IL-6 was a major mechanism through which stress exacerbated respiratory illness symptoms (Cohen et al., 1999b). Similarly, and in line with our current results, a recent investigation found an inverse association between positive emotional style and objective and subjective illness symptoms following rhinovirus exposure in healthy adults, and showed that this relationship was primarily driven by lower nasal levels of IL-6 (Doyle et al., 2006).
Circulating IL-6 concentrations are also elevated in some patients with clinical depression, and correlate with depressive symptoms in healthy individuals and patients with inflammatory disease (Irwin and Miller, 2007). Endotoxin-induced elevations in plasma IL-6 correlate with infection-related symptoms of anxiety and depression in healthy volunteers (Reichenberg et al., 2001; Wright et al., 2005), and there is substantial evidence that peripheral cytokines can signal to the brain to influence mood by altering neuroendocrine and neurotransmitter activity (Dantzer et al., 2008). In the current investigation, optimists had smaller increases in total negative mood during the testing session, an effect primarily driven by lower levels of tension-anxiety and larger increases in mental vigour in optimistic participants. This finding is consistent with the well-established relationship between optimism and improved psychological well-being (Scheier and Carver, 1992), and suggests that smaller IL-6 responses in optimists could be a mechanism protecting against stress-related depressive illness. However, the observed associations between optimism and mood were only significant in the sample as a whole, and we could not distinguish effects among the four experimental groups. Furthermore, in the current study there was no direct association between IL-6 responses and negative mood. We are not sure of the reason for this, as we have previously found an association between IL-6 responses to typhoid vaccine and negative mood in a separate sample of healthy volunteers (Wright et al., 2005). However, the present study had a more complex design and it is possible that a larger sample size would be required to see these effects.
The other main study finding was that in participants exposed to both typhoid vaccine and acute psychological stress, there was a strong positive relationship between optimism and anti-Vi antibody responses, suggesting that acute stress enhanced the antibody response to typhoid vaccine in optimistic individuals. Acute psychological or physical stress at the time of vaccination has been shown to boost antibody responses to a number of other vaccines in humans. In healthy university students, time-pressured mental arithmetic or exercise stress administered prior to vaccination was found to improve antibody responses to influenza vaccine in women (Edwards et al., 2006, 2007), and a recent study showed that antibody responses to Meningococcal A vaccine were enhanced by exercise and mental stress in men (Edwards et al., 2008). Stress-induced enhancement of the immune response to infection is also observed in animals, and is considered an evolutionary adaptive mechanism designed to promote survival during times of heightened adversity such as wounding due to a predator-prey encounter (Dhabhar, 2002). In the present investigation, we did not see any effect of acute stress on antibody responses in our overall sample, but rather only in more optimistic individuals. Edwards and colleagues noted that the adjuvant effects of acute stress were most apparent when the immune response of control participants was relatively poor (Edwards et al., 2006, 2007, 2008), and our control (vaccine/rest) group mounted a satisfactory antibody response to typhoid vaccine. Furthermore, the design of our study was somewhat different, in that acute mental stress was administered after rather than prior to vaccination. However, it is also likely that personality plays a role in the relationship between vaccination, immunity and stress. Trait positive affect has been related to enhanced antibody responses to hepatitis B vaccination in graduate students (Marsland et al., 2006), and a recent investigation showed that elevated levels of dispositional optimism during cardiovascular exercise training were positively associated with training-induced improvements in a primary immune response to a novel antigen in older adults (Grant et al., 2008).
The pathways mediating the relationship between optimism and stress-induced immune changes are unclear. As reported previously (Brydon et al., 2009), tasks induced significant increases in participants’ levels of subjective stress, systolic BP and diastolic BP, with no difference between vaccine and placebo conditions. Participants’ salivary cortisol levels decreased on average over time, with no group differences. However, there were large individual differences in this response, with some participants showing increases in cortisol following tasks (Brydon et al., 2009). There were no significant associations between optimism and subjective ratings of tasks, indicating that optimists did not appraise or engage in the tasks differently from their counterparts. However, in the vaccine/stress group, optimists had lower diastolic BP at 1 and 2 h post-stress, indicative of a faster recovery in autonomic balance, and in both vaccine groups, optimists had lower post-stress levels of salivary cortisol. Catecholamines stimulate IL-6 production from adipocytes and LPS-stimulated monocytes, through binding to β-adrenergic receptors expressed on these cells (Johnson et al., 2005; Mohamed-Ali et al., 2001). Thus, improved recovery in autonomic nervous activity could potentially mediate the smaller IL-6 responses in optimists. Similarly, excess cortisol exposure has been associated with glucocorticoid resistance and dysregulated immunity (Bauer, 2005; Miller et al., 2002), and it is conceivable that optimism could influence immune responses by attenuating stress-induced changes in cortisol. In addition, IL-6 plays a key role in linking innate and adaptive immunity (Naugler and Karin, 2008), and there is evidence that it can promote antibody responses to certain pathogens. For example, co-administration of the IL-6 gene with DNA-based influenza vaccine completely protected mice from a subsequently lethal challenge with the virus (Lee et al., 1999) and stress-induced increases in IL-6 measured at the time of influenza vaccination were a significant predictor of antibody responses to the vaccine in women (Edwards et al., 2006). However, other studies found no association between IL-6 stress responses and antibody responses to other types of vaccine in humans (Edwards et al., 2008), and in the current investigation IL-6 responses were unrelated to anti-Vi antibody measures (data not shown). Furthermore, optimistic individuals had smaller IL-6 stress responses yet larger anti-Vi antibody responses in the vaccine/stress condition. These findings suggest that optimism may influence these immune molecules though distinct pathways.
This investigation was carried out in a sample of healthy male university students, and results may not generalize to other populations. Findings represent secondary analyses of a study with a relatively small number of participants per group that was not originally designed to test the primary hypothesis i.e. that optimism protects against the harmful effects of stress. It is thus conceivable that a larger sample size, including participants with a wider range of dispositional optimism scores, may have generated more striking results. We did not include a coping measure, and it would be useful to know whether coping style mediated the beneficial effects of optimism on stress responses. The cross-sectional design limits conclusions about causality, and it should be noted that peripheral cytokines can feedback to the central nervous system to influence cognitive and affective function (Dantzer et al., 2008), so that IL-6 could potentially affect optimistic mood. Supporting this, higher daily production of nasal mucus IL-6 in healthy adults exposed to rhinovirus or influenza was associated with a decline in concurrent daily positive affect but unrelated to changes in negative affect (Janicki-Deverts et al., 2007). Nevertheless, our findings suggest that trait optimism may be protective for physical and mental health by counteracting stress-induced increases in inflammation and boosting the adjuvant effects of acute stress. These mechanisms may help to explain why people with ‘a generalized tendency to look on the bright side of things’ live a longer and healthier life.
This study was supported by the British Heart Foundation. We are grateful to Dr. Daisy Whitehead for her involvement in data collection, and Drs. Akira Tsuda, Hisayoshi Okamura and Jumpei Yajima for assistance with cortisol measures. We also thank Dr. Linda Perkins-Porras, Dr. Mimi Bhattacharyya, Bev Milne and Bev Murray for their help with vaccination.