Effects of acute systemic inflammation on the interplay between sad mood and affective cognition

Abstract

Experimental endotoxemia is a translational model to study inflammatory mechanisms involved in the pathophysiology of mood disorders including depression. Disturbed affective cognition constitutes a core aspect in depression, but has never been studied in the context of inflammation. We combined experimental endotoxemia with an established experimental mood induction procedure to assess the interaction between acute inflammation and sad mood and their effects on affective cognition. In this randomized cross-over study, N = 15 healthy males received endotoxin (0.8 ng/kg lipopolysaccharide iv) on one study day and placebo an otherwise identical study day. The affective Go/Nogo task was conducted after experimental induction of neutral and sad mood. Inflammatory markers were assessed hourly. Endotoxin application induced a transient systemic inflammation, characterized by increased leukocyte counts, TNF-alpha and interleukin-6 plasma concentrations (all p < 0.01, interaction effects). Mood induction led to greater sadness ratings, with highest ratings when sad mood was induced during inflammation (p < 0.05, interaction effect). Based on a 2 (endotoxin vs. placebo) × 2 (sad vs. neutral mood) × 2 (sad vs. happy Go/Nogo target words) factorial design, we observed a significant target × endotoxin condition interaction (p < 0.01) reflecting slower responses to sad targets during endotoxemia. Additionally, we found a valence × mood interaction (p < 0.05), reflecting slower reaction times to sad targets in sad mood. In summary, acute inflammation and sad mood are risk factors for disturbed affective cognition. The results may reflect a mood-congruency effect, with prolonged and sustained processing of mood-congruent information during acute inflammation, which may contribute to depression risk.

Fig. 1. Effects of endotoxin

WBC counts a, plasma TNF-alpha b, IL-6 c, body temperature d, as well as euthymia e, dysthymia f, and physical sickness symptoms g were repeatedly measured at baseline and up to 6 h post injection of endotoxin (LPS, 0.8 ng/kg body weight) or saline (placebo). LPS administration led to significant increases in WBC counts, plasma IL-6, salivary cortisol, body temperature, indicating a systemic immune activation, as well as to reduced euthymia and increases in dysthymia and the number and intensity of physical sickness symptoms. *p < 0.05, **p < 0.01, ***p < 0.001; results of Bonferroni-corrected post-hoc paired t-tests. For ANOVA results, see text. STADI State Trait Anxiety and Depression Inventory; GASE General-Assessment-of-Side-Effects questionnaire

Fig. 2. Effects of mood induction procedure on VAS mood ratings

Sadness was measured with a VAS before (pre) and after (post) the experimental induction of sad or neutral mood during the endotoxin and the placebo condition. The induction of sad mood led to significant increases in sadness ratings both within the endotoxin and the placebo condition (*p < 0.05, ***p < 0.001, results of Bonferroni-corrected post-hoc paired t-tests). In addition, participants showed higher baseline sadness ratings (i.e., before the mood induction paradigm started) within the endotoxin condition (⁺⁺ p < 0.01, ⁺⁺⁺ p < 0.001, results of Bonferroni-corrected post-hoc paired t-tests), reflecting endotoxin effects on sadness ratings. For ANOVA results, please see text

Fig. 3. Effects of endotoxin and negative mood on reaction time in the affective Go/Nogo task

Delta reaction times for positive vs. negative target words were computed according to Murphy et al. with positive scores indicating a tendency to respond slower and negative scores indicating a tendency to respond faster to sad targets. Differences in delta reaction time between the sad and neutral mood condition were significant only in the LPS condition (*p < 0.05, result of post-hoc paired t-test). For results of ANOVA, please see text