Effect of homozygous deletions at 22q13.1 on alcohol dependence severity and cue-elicited BOLD response in the precuneus

Abstract

Copy number variations (CNVs) can alter the DNA sequence in blocks ranging from kilobases to megabases, involving more total nucleotides than single nucleotide polymorphisms. Yet, its impact in humans is far from fully understood. In this study, we investigate the relationship of genome-wide CNVs with brain function elicited by an alcohol cue in 300 participants with alcohol use disorders. First, we extracted refined neurobiological phenotypes, the brain responses to an alcohol cue versus a juice cue in the precuneus, thalamus and anterior cingulate cortex (ACC). Then, we correlated the CNVs with incidence frequency > 1% to the neurobiological phenotypes. One CNV region at 22q13.1 was identified to be associated with alcohol dependence severity and the brain response to alcohol cues. Specifically, the 22k base-pair homozygous deletion at 22q13.1 affects genes APOBEC3a and APOBEC3b. Carriers of this homozygous deletion show a significantly higher score in the alcohol dependence severity (P < 0.05) and increased response to alcohol cues in the precuneus (P < 10(-12) ) than other participants. Tests of a mediation model indicate that the precuneus mediates the association between the homozygous deletions and alcohol dependence severity. Interestingly, the precuneus is not only anatomically and functionally connected to the ACC and thalamus (the main active regions to the alcohol cue), but also has the most predictive power to the alcohol dependence severity. These findings suggest that the homozygous deletion at 22q13.1 may have an important impact on the function of the precuneus with downstream implications for alcohol dependence.

Figure 1

The mediation model among CNV, brain function and behavior. In the mediation model, CNV is the predictor, brain function is the mediator and alcohol depedence level is the outcome varaible. The mediation effect is measured by a×b, and the direct effect from the predictor to outcome is measured by c. Our data show that brain funcion in the precuneus (yellow highlighted region) mediates the effect from homozygous deletion at 22q13.1 to the alcohol dependence severity, where 7% of variance is explained through mediator, and 1% is directly from the CNV.

Figure 2

Alcohol use phenotype and neurobiological phenotype. a) The main alcohol dependence severity level comprising mainly ADS-con, AUDIT total score, ICS-fc and ICS-pc. b) Brain regions are significantly, and positively correlated to the alcohol dependence severity, passing 0.05 FDR control. The top three regions are precuneus, thalamus and ACC, ordered by the correlation strength. The activation of these regions comprises the neurobiological phenotypes.

Figure 3

The CNVs at 22q13.1 and their association with brain function and alcohol dependence severity. a) The distribution of the CNVs in this region includes 6 homozygous deletions in brown, 5 single deletions in red, 5 insertions in green and 289 neutral in blue. This region codes for many genes close to one other, the main affected gene is CTH4. b) The brain regions correlated with the CNVs in this region, mostly in the precuneus with the minimum p value < 1E-12. c) The modulation of homozygous deletion status on alcohol dependence severity level. For each box, the central mark is the median, the edges of the box are the 25th and 75th percentiles, the whiskers extend to the most extreme data points not considered outliers, and outliers are plotted individually as cross markers.