Nicotine Abstinence Influences the Calculation of Salience in Discrete Insular Circuits

Abstract

Background: Insular subdivisions show distinct patterns of resting-state functional connectivity (rsFC) with specific brain regions, each with different functional significance. Seeds in these subdivisions are employed to characterize the effects of acute nicotine abstinence on rsFC between insula subdivisions and brain networks implicated in addiction and attentional control.

Methods: In a within-subjects design, resting-state blood oxygen level-dependent data were collected from treatment-seeking smokers (N= 20) following smoking satiety and again following 48 hours of nicotine abstinence. Three right hemisphere insular regions of interest (dorsal, ventral, and posterior) served as seeds for analyses. Indices of both static and dynamic rsFC were obtained and correlated with indices of subjective withdrawal and behavioral performance.

Results: Abstinence-induced physiological, subjective, and cognitive differences were observed. Overall dynamic rsFC was reduced during abstinence, and circuits containing each insular seed showed changes in rsFC as a function of nicotine abstinence. Specifically, dorsal and posterior insular connections to the default mode and salience networks were enhanced, while a previously undescribed ventral insular connection to the executive control network was reduced. Further, static rsFC was significantly correlated with subjective ratings of aversive affect and withdrawal in the modified ventral and posterior insular-seeded circuits.

Conclusions: As predicted, divergent connections between insula subdivisions and anticorrelated resting brain networks were observed during abstinence. These changes reflect an attentional bias toward aversive affective processing and not directly away from exogenous cognitive processing, suggesting a coordinated modulation of circuits associated with interoceptive and affective processing that instantiates an aversive state during nicotine abstinence.

Keywords: Abstinence; Affect; Cessation; Functional connectivity; Insula; Nicotine.

Figure 1. Seeds and Targets

Clusters in Dorsal, Ventral and Posterior insula served as seeds for static rsFC (see Table S2 for coordinates) with targets comprised of group ICA-derived voxel maps of each of three large-scale networks relevant to attention and addiction: Executive Control Network (ECN), Salience Network (SN), and Default Mode Network (DMN). All rsFC comparisons were corrected for multiple comparisons.

Figure 2. s-rsFC results

Maps of circuits showing a significant SESSION effect (left) and extracted s-rsFC values from identified clusters during satiety (green) and abstinence (red) in A) VEN (purple), B) POS (orange), and C) DOR (yellow) insula. Bar plot of the observed effect is included to aid interpretation.

Figure 3. Dynamic rsFC results

Markov matrices describing transitions between identified states during satiety (top) and abstinence (bottom). P(remain in state) calculated as sum of diagonal within each matrix. A significant effect of SESSION on P(remain in state) is observed.

Figure 4. rsfc Correlations

Significant correlations between s-rsFC and WSWS measures are observed between A) POS-IFC and WSWS sadness (1) and POS-dACC circuit and WSWS anger (2) during satiety, and between B) VEN-dlPFC circuit and WSWS Craving during abstinence