Inflammatory Challenge in Human Aggression

Project Summary: While elevations of circulating pro-inflammatory modulators correlate directly with variables of aggression, and direct application of cytokines to specific cortico-limbic regions in animals elicit aggressive responding, no studies have tested the hypothesis that acute increases in pro-inflammatory modulators can/will

increase aggressive behavior in humans. We aim to demonstrate a causal relationship between pro- inflammatory cytokines and aggression in human subjects by showing that an acute pro-inflammatory state, via endotoxin challenge, will increase aggressive responding, anger ratings, and hostile social cognition, to a greater

degree in “aggressive” (n = 45), compared with “non-aggressive” (n = 45), individuals with mood/anxiety/ stress- related and/or personality disorders. The proposed study is a double-blind comparison of endotoxin/placebo challenge in the same individuals (within-subject) as a function of aggression status. Aggressive individuals will

have high lifetime aggression (> 12 on Life History of Aggression: LHA) and be positive for an average of two anger attacks per week and/or three anger attacks per year that include physical assault of another person and/or or non-trivial destruction of property. “Non-Aggressive” individuals will be similar diagnostically but will

have low lifetime aggression. “Aggressive responding” will be assessed using the Taylor Aggression Paradigm (TAP), “Anger” will be assessed by self-reported assessments (POMS), and “Hostile Social Cognition” will be assessed by the Video-SEIP (V-SEIP) paradigm. The primary plasma pro-inflammatory outcome measures will

be a composite of CRP, IL-6, IL-8, and TNF-α (as in our previous studies). MRI scans will including task-based scans involving “explicit” and “ambiguous” social threat. We hypothesize that aggressive responding, anger ratings, and hostile social cognition (Primary Outcomes) and Composite Plasma Pro-Inflammatory Marker levels

(Secondary Outcome), will be greater after endotoxin, compared with placebo, and will be greater in “aggressive” than “non-aggressive” study participants. We also hypothesize that these variables will correlate with dimensional measures of aggression. In addition, we hypothesize that amygdala responses to explicit social

threat (anger faces) will be enhanced (greater BOLD fMRI signal response) after Endotoxin, and that cortico- limbic responses to ambiguous social threat (V-SEIP) will be reduced (i.e., lesser BOLD fMRI signal response)

in all study participants but reduced to a greater extent in “aggressive”, compared with “non-aggressive”, study participants. Finally, we hypothesize that the pro-inflammatory effects of the endotoxin challenge will result in reduced connectivity between the functional edges supporting higher aggressive behavior and that endotoxin

challenge will facilitate stronger connections among nodes associated with low aggressive behavior. If supported, this study will provide a strong rationale for clinical trials of anti-inflammatory agents in impulsive aggressive individuals.