Early Life Stress Moderates the Relation Between Inflammation and Nucleus Accumbens Gray Matter Volume in Adolescents

Early life stress (ELS) has been associated with increased development of psychopathology, like depression, in late adolescence and early adulthood. ELS has also been associated with heightened inflammation and anomalies in reward processing. The neuroimmune network hypothesis offers a framework to model the relations among ELS, inflammation, and the brain, suggesting that there is bidirectional communication between the immune system and brain regions associated with reward, and that exposure to stressors alters these neuroimmune relations. In this study, we investigated how deprivation, as a dimension of ELS, affects the relation between the inflammation marker C-Reactive Protein (CRP) and gray matter volume (GMV) of the nucleus accumbens (NAcc), a brain region associated with reward.
 
Our sample included 110 (63F/47M) individuals (15.76 ± 1.27 years) who were participating in a longitudinal study focused on long-term outcomes of ELS. We computed ELS deprivation scores, a composite of neighborhood level socio-economic metrics and the Multidimensional Neglectful Behavior Scale (MNBS). CRP was used as a marker representing inflammation levels, and GMV of the NAcc was measured using a 3T MRI scanner via T1w images, which were processed through the cat12 SPM pipeline. Using R, stepwise regression was conducted to assess relations among the primary variables and to investigate the moderating effect of ELS disadvantage on the relation between CRP and NAcc GMV. Covariates of interest including age, sex, assay batch, BMI, and scanner upgrade were controlled in the statistical analyses. 
 
We found a significant interaction (β[95%CI]= -0.303, t=-2.40, p=0.0183 ) in which ELS moderated the relation between CRP and NAcc GMV. The relation between inflammation and GMV differed between individuals with high versus low and average levels of ELS. Using standard deviation cutoffs to divide the sample into high, average, and low ELS groups, we found that the high ELS group showed a negative linear relation between CRP levels and NAcc GMV. For the high ELS group, a high inflammation value was associated with lower NAcc GMV. We also conducted an exploratory analysis of longitudinal effects of ELS on the relation between the change in CRP and the change in NAcc GMV over approximately two years with a subset of the sample (n = 48) who had two timepoints of inflammation and scan data. No significant longitudinal interaction was observed in this exploratory longitudinal interaction model.
 
Our cross-sectional findings support a growing body of research linking CRP to alterations in reward processing regions of the brain. Most previous studies have focused on functional Magnetic Resonance Imaging in examining neural responses to reward stimuli. Our structural findings add to this body of work, given that brain structure is posited to underlie function. The moderating effect of ELS on the association between CRP and NAcc GMV support predictions from the neuroimmune network hypothesis and indicates that childhood deprivation can affect the relation between inflammation and brain structure in reward processing regions.