Work

Prevention of Stress-Provoked Endothelial Injury by Values Affirmation: a Proof of Principle Study

Year: 2015

Type: article

Abstract: The link between stress and disease has been of paramount interest to researchers and clinicians for decades. Hans Selye was among the first to recognize the effect of stress on disease relevant processes [1], and in tandem, Paul MacLean noted the associations among emotion, the brain, and health [2]. To date, some of the strongest associations have been established between stress and cardiovascular disease (e.g., Dimsdale [3]). For example, psychological stress can provoke acute cardiac events [4] and induce transient myocardial ischemia [5], and a number of epidemiological studies have linked anger, hostility, and chronic perceived stress to cardiovascular disease mortality (e.g., Haynes, Feinleib, and Kannel [6]; Rosengren et al. [7]). Critical to cardiovascular disease is the health of vascular endothelial cells, which play an essential role in maintaining vascular tone and the integrity of blood vessels. Endothelial injury causes endothelial cells to lose integrity, progress to senescence and detach into the circulation [8, 9]. Endothelial cell-derived microparticles (EMPs) are phospholipid rich, submicron particles derived and released from the membranes of activated or apoptotic endothelial cells [8, 10], and they can serve as markers of endothelial cell injury in humans. EMPs modulate inflammation via leukocyte activation and transendothelial migration and have procoagulant activity [11]. Thus, EMPs may have a direct role in atherosclerosis-related cardiovascular disease onset. EMPs may also contribute to other vascular disorders, including stroke [12] and preeclampsia [13]. Although many studies have investigated stress effects on health-related physiology (e.g., cortisol [14], immune [15], and autonomic [16] responses), few studies have tested whether psychological treatments can influence cellular processes directly relevant for disease. EMPs are a particularly important target in that regard; stress-induced EMP activity represents one plausible mechanism underlying the association between psychological stress and cardiovascular disease. In support of this hypothesis, experimentally provoked anger has previously been shown to elevate circulating EMPs [17], suggesting that psychological stress can cause endothelial cell injury. If psychological stressors can impact cellular processes relevant for cardiovascular disease risk, then psychological processes may be able to prevent them, and demonstrating such effects could lead to enhanced understanding and clinical deployment of psychological interventions. However, no prior research has investigated whether endothelial cell injury can be mitigated by a psychological inoculation. Among such procedures, values affirmation has emerged as a practice that is both effective at reducing stress—including cortisol reactivity [18], catecholamine secretion [19], and performance improvement [20–23]—and deployable on a large scale in diverse populations [20, 21, 23]. Here, we present a proof of principle study that tested whether social evaluative threat (SET) and values affirmation would have opposing effects on markers of endothelial cell injury. We recruited healthy participants who were randomized to one of three conditions: SET Alone, SET preceded by a buffering values affirmation task (SET + Values Affirmation), or a non-stressful Control task. Blood was sampled immediately prior to the implementation of the SET or Control task, immediately post-SET/Control task (Post 1) and approximately 25 or 40 min later (Post 2) (Fig. 1). Circulating EMPs were assessed by flow cytometry as previously described [24, 25]. EMPs expressing CD62E (CD62E+ EMPs) reflect endothelial cell activation, EMPs expressing CD31 (CD31+ EMPs) reflect endothelial cell apoptosis, and EMPs expressing CD51 (CD51+ EMPs) reflect both endothelial cell activation and apoptosis. Study procedure. Participants were randomized to one of three groups after which they completed two tasks, one in the Inoculation phase and one in the Stressor phase. Blood was drawn at three time points including Pre-Trier Social Stress Test, Post 1 and Post 2. SET social evaluative threat; m minutes Those assigned to the SET + Values Affirmation condition completed a self-affirmation of values task immediately prior to SET, which consisted of reflecting on and writing about aspects of one's life for which one is grateful, including family and social support, social roles and important values, or prized skills [26, 27]. A key ingredient of values affirmation is thought to be affirmation of overall perceptions of personal self-worth, efficacy, and adequacy. It is thought to bolster internal resources for coping with threat, thus reducing threat appraisal when faced with a challenging situation [28]. Prior research has shown that values affirmation attenuates stress and defensiveness and improves performance among those contending with racial [20, 21] and gender identity [23] threat. Participants assigned to the SET + Values Affirmation group wrote about personally important values, whereas those assigned to SET Alone and Control groups wrote about values least important to them but important to others; this was not expected to confer benefit. SET was induced via the well-established Trier Social Stress Test [14, 29], a task in which participants are instructed to prepare a public speech and perform mental math problems before a panel of “expert judges” (here, two paid actors). The Trier Social Stress Test reliably increases sympathetic activity and reduces parasympathetic cardiac control [30] via prefrontal-brainstem connections [31] and is similar to other procedures found to influence vascular endothelial function [32, 33]. The Control group performed a task that involved planning and cognition with no evaluative threat. Participants were healthy, English-speaking adults (N = 32) (23 male; mean age (SD) = 22.6 (3.8) years) randomized into one of three groups: SET Alone (n = 7; 4 male), SET + Values Affirmation (n = 11; 9 male) and Control (n = 14; 10 male). Exclusion criteria relevant to the current study included personal history of cardiovascular disease, use of anti-hypertensive medication and smoking. Three participants from the SET + Values Affirmation group missed one blood draw each due to random equipment failure, leaving full data sets available from 29 participants (SET Alone (n = 7), SET + Values Affirmation (n = 8), and Control (n = 14)) (Fig. 1). Data were not complete for one participant in the SET Alone group for cortisol (n = 6) and one participant in the Control group for the psychological stress measure (n = 13). In the SET + Values Affirmation group, three participants were added to the analysis of cortisol (n = 11), and two were added to the analysis of psychological stress (n = 10). All participants gave written informed consent, and study procedures were approved by the Institutional Review Board of Columbia University and were carried out in accordance with the provisions of the World Medical Association Declaration of Helsinki. Participants reported to the laboratory between 1 and 2 pm, having refrained from consuming caffeine, eating a large meal, and participating in any form of exercise for 2 h prior to the session and from consuming alcohol and participating in heavy exercise for 48 h prior to the session. Also, participants obtained a normal night's sleep on the preceding night. Participant compliance was confirmed with self-report. Upon arrival, an indwelling catheter (21-gauge) was inserted into the non-dominant forearm. Next, participants completed the values affirmation task. Blood samples (4.5 mL) were drawn into citrated tubes 70 min after catheter insertion and a resting baseline condition, at the end of the SET/Control condition, and then 25–40 min after completion of additional cognitive tasks (Fig. 1). At the end of the study, participants were fully debriefed and paid. The values affirmation task was designed to have two versions: self-focused and other-focused. In the self-focused task, participants were instructed to select one of the following six values that was most important to them: athletic ability, creativity, relationship with family and friends, religious values, sense of humor, music, or art. They then wrote a short description and answered brief questions concerning why that value was important to them. In the other-focused task, participants were presented with the same list of values, but instead, were instructed to select one that was least important to them. They then wrote a short description and answered brief questions concerning why that value might be important to someone else, with the brief questions being the same as the self-focused task, but reoriented toward other people. Two confederates (paid actors) wearing white lab coats were introduced to participants as professors. The confederates informed participants that they would be required to give a 5-min speech, with 3 min to prepare the speech. Participants were told that the topic of the speech would be revealed shortly. At the onset of the speech preparation period, the speech topic—describing their strengths and weaknesses as a candidate for their dream job—was displayed on a computer screen. The job was selected from a questionnaire completed previously. At the end of the 3-min speech preparation period, a conspicuous video camera was turned on, and participants were instructed to begin delivery of their speeches to the confederates. Throughout speech delivery, the confederates used neutral facial expressions and frequently interrupted the participants with reminders to adhere to instructions. Following speech delivery, the confederates administered a serial subtraction task, instructing participants to count backwards by 13s from the number 1022 and to start over when mistakes were made. In this task, the experimenter informed participants that they would have a few minutes to imagine a day spent with a friend who was visiting New York City, and following this period, they would be asked to describe the day verbally to the experimenter. Participants were told that there were no correct answers, and all instructions were designed to minimize social threat. Equivalent time was allotted for the speech preparation and delivery periods as in the Trier Social Stress Test. Flow cytometry was used to differentiate EMP phenotypic surface marker expression associated with endothelial cell activation (CD62+ EMPs), apoptosis (CD31+ EMPs), or both (CD52+ EMPs), based on a technique of Jimenez and colleagues [34]. Sample preparation occurred within 2 h of each blood draw. Citrated blood was centrifuged at 160×g for 10 min to prepare platelet-rich plasma, and the platelet-rich plasma was further centrifuged for 8 min at 1000×g to obtain platelet-poor plasma, and 50 μL of the platelet-poor plasma was then incubated with 5 μL of phycoerythrin-conjugated monoclonal antibody to CD62E (BD Biosciences). This antigen is specific for endothelial cells, obviating the need for a second antibody labeling. Unlike CD62E, the expression of CD31 and CD51 occurs on both platelet microparticles (PMPs) and EMPs. Thus, as CD42b is only present on platelets, fluorescein isothiocyanate-labeled anti-CD42b was additionally used to distinguish between PMPs and EMPs expressing CD31 and separately EMPs expressing CD51. As such, 50 μL of platelet-poor plasma was incubated with 4 μL of phycoerythrin-conjugated monoclonal antibody to CD31 (AbD Serotec) and separately CD51 (AbD Serotec), along with 4 μL of fluorescein isothiocyanate-conjugated monoclonal antibody to CD42b (AbD Serotec). After incubation, 1 mL of 0 × 01 M PBS buffer was added, and the samples were placed in polypropylene tubes for analysis by flow cytometry (BD FACSCalibur) at a medium flow rate for a 30-s period. EMPs are defined as the number of particles with size <1 × 5 μm, which are positively labeled by CD62E (CD62E+ EMPs), positively labeled by CD31 and negatively labeled by CD42 (CD31+ EMPs), and positively labeled by CD51 and negatively labeled by CD42 (CD51+ EMPs). For all experiments, appropriate fluorescein isothiocyanate- and phycoerythrin-labeled isotype-matched IgG were also used to determine non-specific binding. Using standard beads, total flow cytometry counts for each experiment were converted to the number of EMPs per microliters as previously described. Participants provided saliva samples and psychological stress ratings (i.e., “How stressful was the task you just completed?”) on a Likert scale of 0 to 8 over seven time points during the session. The three time points coinciding with EMP collection will be presented here. Saliva samples, collected with Salivettes, were sent to the Kirschbaum Laboratory (Dresden, Germany) to be assayed for free circulating cortisol (nmol/L). Total counts (counts/μL) of each measure (CD31+ EMPs, CD51+ EMPs, and CD62E+ EMPs) were log-transformed due to skewed distributions identified with Kolmogorov-Smirnov tests. A 3 × 3 repeated measures ANOVA was conducted on each EMP marker including the within-subjects factor of TIME (Pre-Trier Social Stress Test, Post 1, Post 2) and the between-subjects factor of GROUP (SET Alone, SET + Values Affirmation, Control). The critical test of these ANOVAs was the interaction of TIME × GROUP. Next, the following planned contrasts were tested on each EMP marker: (1) Post 1 and Pre-Trier Social Stress Test levels within each group were compared using a paired t-test, (2) a change score between Post 1 and Pre-Trier Social Stress Test was calculated, and then this score was compared as a function of group with an independent t-test, (3) baseline effects were assessed by comparing Pre-Trier Social Stress Test values as a function of group with an independent t-test, and (4) group differences were assessed at Post 1, controlling for Pre-Trier Social Stress Test levels by employing multiple linear regression where the outcome was Post 1 values, and the predictors were Group and Pre-Trier Social Stress Test values. These data were submitted to the same ANOVA as above. Post hoc t-tests were conducted accounting for multiple comparisons with Bonferroni correction. As predicted, the interaction of TIME × GROUP from ANOVA was significant for each EMP marker (CD31+ F(4,52) = 4.88 p = 0.002; CD51+ F(4,52) = 3.24, p = 0.019; CD62E+ F(4,52) = 3.89, p = 0.008). Planned contrasts revealed that the SET Alone group demonstrated significantly increased circulating levels of all three EMP populations from Pre-Trier Social Stress Test to Post 1 relative to the Control group (CD31+ t(19) = 3.06, p = 0.006; CD51+ t(19) = 2.74, p = 0.013; CD62E+ t(19) = 3.08, p = 0.006). EMP levels did not change in the Control group (all p > 0.10). In addition, levels of all three EMPs increased significantly less in the SET + Values Affirmation group relative to the SET Alone group (CD31+ t(13) = 2.79, p = 0.015; CD51+ t(13) = 2.81, p = 0.015; CD62E+ t(13) = 2.89, p = 0.013). EMP changes in the SET + Values Affirmation and Control groups were statistically equivalent (all p > 0.10). Thus, values affirmation prevented the endothelial injury caused by SET. There were no significant differences on the three measures of EMPs between groups at Pre-Trier Social Stress Test (see Table 1 and Fig. 2). Averaging Post 1 and Post 2 time points yielded similar results for all EMPs. Results for each planned contrast and endothelial cell-derived microparticle (EMP) marker SET social evaluative threat, STE standard error of the mean, CI confidence interval p < 05; p < 01 Average counts/microliters as a function of time and group for each endothelial cell-derived microparticle (EMP) marker. In general, (a) CD31+ EMPs, (b) CD51+ EMPs, and (c) CD62E+ EMPs increased as a function of the Trier Social Stress Test for the social evaluative threat (SET) Alone group, but did not for the SET + Values Affirmation and Control groups, indicating that (1) SET affects EMPs and (2) a psychological inoculation can block such effects. Error bars represent one standard error of the mean The interaction of TIME × GROUP on cortisol and psychological stress was significant (F(4,56) = 3.54, p = 0.012 and F(4,54) = 3.74, p = 0.009, respectively), as shown in Fig. 3. For cortisol, post hoc t-tests revealed higher values at Post 1 relative to the Pre-Trier Social Stress Test time point in the SET Alone = p = and SET + Values Affirmation = 3.08, p = groups and higher values for the SET Alone group relative to at Post 1 = p = and Post 2 = p = with no other effects being no differences were found among groups at Pre-Trier Social Stress Test, and no was found among the SET + Values Affirmation and Control groups at Post no was found among the SET Alone and SET + Values Affirmation groups at Post these results that the SET + Values Affirmation group in between the other two cortisol and psychological stress measures as a function of time and A was conducted using (a) cortisol and (b) psychological stress measures to support for the of the social evaluative threat (SET) and values affirmation on endothelial cell-derived microparticles In general, cortisol and stress ratings for the SET + Values Affirmation group between those for SET Alone and Control with the of the Post 2 time point where psychological stress ratings for the SET + Values Affirmation group to baseline and those of the Control group increased Results for ratings of psychological stress are shown in Fig. 3. Post hoc t-tests revealed higher values at Post 1 relative to Pre-Trier Social Stress Test = p = and Post 2 = p = in the SET Alone higher values at Post 1 only = p = in the SET + Values Affirmation group, indicating their stress levels to baseline at Post and higher values at Post 2 only = p = for the Control This is not in that the Control group perceived the cognition tasks prior to the Post 2 as stressful relative to the Control task they performed prior to the Post 1 higher values were in the SET Alone Control group at Post 1 = p = and a significant was between the SET + Values Affirmation and Control groups at Post 1 = p = other effects were As with cortisol, stress ratings for the SET + Values Affirmation group between those for SET Alone and This proof of principle study is the first that a psychological designed to psychological threat in populations can prevent endothelial provoked by psychological In to that this form of stress circulating EMPs marker of endothelial and that a brief psychological inoculation designed to the appraisal of stress this these that psychological events to one's social and have and direct on vascular These were by and with cortisol and psychological stress as values affirmation social cognitive processes that in higher to the of of and social and or racial associations that one's as threat been thought to The psychological of SET involved similar to as has been shown to of the important for stress of social and the of threat and SET into directly relevant for The psychological effects on endothelial health we describe may be by autonomic control of vascular performance and This is with other studies of self-affirmation that have shown effects on autonomic and to stressful tasks the effects of stress and self-affirmation to directly relevant for cardiovascular disease risk, the current between psychological and or We some important of this may be asked whether and endothelial injury occurred in Although we use the endothelial injury this free circulating EMPs are as markers of endothelial As for the of the many aspects of the Trier Social Stress Test is to have but to the and not critical for research is to the effects of many such those that lead to changes in whether acute reactivity of endothelial function is also studies that markers of endothelial are associated with an increased of cardiovascular disease and cardiovascular disease events These studies assessed markers of endothelial function at and did not the of stress that each was at the time of the In addition, studies have demonstrated acute effects of a meal, acute mental a endothelial these effects have not been linked to is whether stress-induced endothelial cardiovascular disease and cardiovascular disease study was conducted with healthy and could be that those at for cardiovascular disease have effects. This is important of the values affirmation changes in this The effects of the on physiology are there to be effects on performance and health as a stress inoculation with to cardiovascular health these data are due to the They serve as a proof of a for At the same be noted that within samples, effect can be found in the of the current whereas effect Thus, such significant be with in when for studies However, of effect is (1) multiple and the (2) the by included or on is and (3) the size is and then The current study did not any of these the connections in the current For example, self-affirmation of values in the of is thought to have multiple psychological including and the from the threat and It is that any or all of these psychological processes could have effects on for cardiovascular disease, with multiple This was by the of to and to

Source: Annals of Behavioral Medicine

DOI: 10.1007/s12160-015-9756-6