Adolescent alcohol exposure decreases frontostriatal resting-state functional connectivity in adulthood

Abstract

Connectivity of the prefrontal cortex (PFC) matures through adolescence, coinciding with emergence of adult executive function and top-down inhibitory control over behavior. Alcohol exposure during this critical period of brain maturation may affect development of PFC and frontolimbic connectivity. Adult rats exposed to adolescent intermittent ethanol (AIE; 5 g/kg ethanol, 25 percent v/v in water, intragastrically, 2-day-on, 2-day-off, postnatal day 25-54) or water control underwent resting-state functional MRI to test the hypothesis that AIE induces persistent changes in frontolimbic functional connectivity under baseline and acute alcohol conditions (2 g/kg ethanol or saline, intraperitoneally administered during scanning). Data were acquired on a Bruker 9.4-T MR scanner with rats under dexmedetomidine sedation in combination with isoflurane. Frontolimbic network regions-of-interest for data analysis included PFC [prelimbic (PrL), infralimbic (IL), and orbitofrontal cortex (OFC) portions], nucleus accumbens (NAc), caudate putamen (CPu), dorsal hippocampus, ventral tegmental area, amygdala, and somatosensory forelimb used as a control region. AIE decreased baseline resting-state connectivity between PFC subregions (PrL-IL and IL-OFC) and between PFC-striatal regions (PrL-NAc, IL-CPu, IL-NAc, OFC-CPu, and OFC-NAc). Acute ethanol induced negative blood-oxygen-level-dependent changes within all regions of interest examined, along with significant increases in functional connectivity in control, but not AIE animals. Together, these data support the hypothesis that binge-like adolescent alcohol exposure causes persistent decreases in baseline frontolimbic (particularly frontostriatal) connectivity and alters sensitivity to acute ethanol-induced increases in functional connectivity in adulthood.

Keywords: Sprague-Dawley; adolescence; ethanol; fcMRI; prefrontal cortex; striatum.

Figure 1

(A) Experiment timeline. Animals in Group 1 were exposed to adolescent intermittent ethanol (AIE; 5 g/kg ethanol [EtOH], 25% v/v in water, i.g., 2-day-on, 2-day-off, postnatal day [P]25-P54). Control (CTRL) animals in Group 2 received an equivalent volume of water during the same exposure regimen. In adulthood, each animal underwent two scanning days. (B) Procedure on each scan day. Three scans were acquired in immediate succession: 1) 10-minute baseline resting-state scan [PreChall], 2) 25-minute challenge scan [Chall], and 3) 10-minute post-challenge scan [PostChall]. During the Chall scan, animals were injected i.p. with either 2 g/kg EtOH (20% v/v) or an equivalent volume of saline (SAL) 6 minutes after scan start. Each animal served as its own saline control for acute EtOH challenge, and the order of injection was counterbalanced.

Figure 2

ROIs used for analyses shown on the T2 template. All ROIs were manually drawn by co-registering the T2 template to the Paxinos & Watson rat brain atlas (6^th Ed). Numbers below slices indicate location from bregma.

Figure 3

(A) Functional connectivity matrix analysis of adult baseline PreChall data following AIE is shown collapsed across scan day. Z-score correlation matrices for CTRL and AIE groups, as well as the difference between AIE and CTRL Z-scores among each ROI pair (AIE – CTRL). Red indicates increased connectivity and blue indicates decreased connectivity. AIE adults have significantly less resting-state connectivity between PFC subregions (PrL-IL and IL-OFC), as well as between PFC regions and the striatum (PrL-NAc, IL-CPu, IL-NAc, OFC-CPu and OFC-NAc). Asterisks (*) indicate statistical significance (p < 0.05). (B) Sagittal view of the rat brain showing ROI-pair connectivity that was significantly decreased in AIE adults.

Figure 4

Voxel-wise correlation maps of PreChall data (threshold = 0.5 average Z-score) using the same PFC bilateral ROIs that significantly differed between AIE and CTRL in the matrix analysis. Corresponding to the results from the matrix correlation analysis, AIE rats show significantly less PrL, IL and OFC connectivity than CTRL.

Figure 5

Representative BOLD time course (CPu ROI normalized to the first 20 seconds) during the 25-minute Chall scan. Dashed line indicates injection time-point (minute 6). Compared to SAL, acute EtOH decreased BOLD signal across all ROIs examined, regardless of exposure history (also see Table 1). No effect of AIE was observed.

Figure 6

ROI-pair correlational analysis of data 20 minutes after injection of saline (SAL) or ethanol (EtOH), and the difference between SAL and EtOH (SAL-ETOH). As shown in the ROI-pair correlation matrices, EtOH significantly increased overall connectivity relative to SAL in CTRL (*), but not AIE adults (ns).