You say 'prefrontal cortex' and I say 'anterior cingulate': meta-analysis of spatial overlap in amygdala-to-prefrontal connectivity and internalizing symptomology

Abstract

Connections between the amygdala and medial prefrontal cortex (mPFC) are considered critical for the expression and regulation of emotional behavior. Abnormalities in frontoamygdala circuitry are reported across several internalizing conditions and associated risk factors (for example, childhood trauma), which may underlie the strong phenotypic overlap and co-occurrence of internalizing conditions. However, it is unclear if these findings converge on the same localized areas of mPFC or adjacent anterior cingulate cortex (ACC). Examining 46 resting-state functional connectivity magnetic resonance imaging studies of internalizing conditions or risk factors (for example, early adversity and family history), we conducted an activation likelihood estimation meta-analysis of frontoamygdala circuitry. We included all reported amygdala to frontal coordinate locations that fell within a liberal anatomically defined frontal mask. Peak effects across studies were centered in two focal subareas of the ACC: pregenual (pgACC) and subgenual (sgACC). Using publicly available maps and databases of healthy individuals, we found that observed subareas have unique connectivity profiles, patterns of neural co-activation across a range of neuropsychological tasks, and distribution of tasks spanning various behavioral domains within peak regions, also known as 'functional fingerprints'. These results suggest disruptions in unique amygdala-ACC subcircuits across internalizing, genetic and environmental risk studies. Based on functional characterizations and the studies contributing to each peak, observed amygdala-ACC subcircuits may reflect separate transdiagnostic neural signatures. In particular, they may reflect common neurobiological substrates involved in developmental risk (sgACC), or the broad expression of emotional psychopathology (pgACC) across disease boundaries.

Figure 1

Study selection. Number of studies is given in bold letters.

Figure 2

Converging evidence of disrupted amygdala functional connectivity with two separate ACC subregions across 46 internalizing, genetic and environmental risk studies. Results of coordinate-based meta-analysis that included 2401 individuals. ACC, anterior cingulate cortex; ALE, activation likelihood estimation; pgACC, pregenual ACC; RSFC, resting-state functional connectivity; sgACC, subgenual ACC.

Figure 3

ACC meta-analytic peaks show unique patterns resting-state functional connectivity (FC) (a) and task-related co-activation (b), suggesting unique subcircuits. (a) Resting-state FC data in 1000 healthy individuals generated via www.Neurosynth.org, P<0.01 FDR corrected. (b) Coordinate-based meta-analysis of areas that co-activate with ACC meta-analytic peaks. A total of 971 healthy individuals contributed to pgACC and 493 to sgACC. Thresholded with cluster-level FWE correction P<0.05 and voxel-level, P<0.001. ACC, anterior cingulate cortex; FDR, false discovery rate; FWE, family-wise error; pgACC, pregenual ACC; sgACC, subgenual ACC.

Figure 4

ACC meta-analytic peaks show unique functional fingerprints. Behavioral domains (number of studies) associated with activity in each ACC peak, according to studies in the www.BrainMap.org database (accessed on 2 May 2016). Behavioral domains with <25 corresponding studies are not included. A total of 971 healthy individuals contributed to pgACC and 493 to sgACC. ACC, anterior cingulate cortex; pgACC, pregenual ACC; sgACC, subgenual ACC.