Differential gene expression linked to alcohol use disorder, offering new treatment possibilities

New study reveals gene expression differences in brain regions linked to addiction, highlighting avenues for innovative treatments for alcohol use disorder and opportunities for drug repurposing.

Study: Gene expression differences associated with alcohol use disorder in the human brain. Image credit: Roman Zaiets / Shutterstock

A recent study in the journal Molecular Psychiatry provided neurobiological insights into AUD by exploring the meta-analyzed gene expression pattern in two addiction-relevant brain regions, namely the nucleus accumbens (NAc) and the dorsolateral prefrontal cortex (DLPFC).

By conducting meta-analyses across multiple independent data sets, the study identified differentially expressed genes (DEGs) linked to AUD, providing robust findings due to increased statistical power and large size of sample

Meta-analyses revealed a total of 476 DEGs, with 25 overlapping between NAc and PFC, highlighting shared and region-specific gene expression patterns associated with AUD.

Prevalence and neurological knowledge about AUD

Millions of deaths occur each year due to alcohol abuse. Although several genome-based studies have indicated the heritable nature of AUD, the gene regulatory landscape related to this disorder has remained unclear. Understanding the neurobiological mechanisms should help identify a potential target for developing effective interventions to alleviate AUD.

The NAc, prefrontal cortex (PFC), and DLPFC regions of the brain are associated with reward and addiction pathways as components of the dopaminergic mesolimbic system. These brain regions are closely related to addiction; for example, NAc is associated with the binge/intoxication stage, and DLPFC involves the worry/anticipation stage.

The PFC regulates dopamine release in the NAc. Several studies have shown that impairment of the PFC negatively affects executive function and impulsivity and increases involvement in risky behaviors. Taken together, the NAc and PFC brain regions are highly connected to AUD.

A limited number of studies have explored AUD-related bulk RNA-seq gene expression in the human brain. This study’s use of meta-analysis on independent datasets significantly strengthens the reliability of the findings. These studies enabled the identification of differential gene expression (DGE) in the brains of patients with AUD.

About the study

Postmortem human NAc and DLPFC samples were obtained from 122 candidates, i.e., 61 AUD and 61 non-AUD, as part of the Lieber Institute for Brain Development (LIBD) Human Brain Repository.

AUD cases and controls were determined based on Diagnostic and Statistical Manual of Mental Disorders-5th Edition (DSM-5) symptoms. AUD cases were those who developed more than two symptoms within twelve months, while non-AUD controls were those with no lifetime history of DSM-5 AUD symptoms. In addition, non-AUD cases showed postmortem ethanol toxicology below 0.06 g/dL.

AUD cases and controls were matched for major depressive disorder (MDD) and smoking status. It should be noted that MDD and smoking are the two most frequent comorbidities of AUD.

RNA was extracted from AUD and non-AUD tissues, and Illumina TruSeq Total RNA Stranded RiboZero Gold was used for library preparation. These samples were named the NAc_LIBD and PFC_LIBD datasets. Other samples obtained from UT Austin and NYGC were named NAc_UT, PFC_UT, and PFC_NYGC, respectively.

All RNA-seq data from different sources were processed using various bioinformatics tools, such as Trimmomatic transcriptome and GENCODE v40 (GRCh38), and quality control (QC) metrics were calculated. The proportion of different cell types including microglia, macrophages, excitatory neurons, oligodendrocyte precursor cells (OPCs), GABAergic neurons, oligodendrocytes, T cells, astrocytes, and medium spiny neurons (MSNs) was estimated ), for PFC and NAc.

Linear regression analysis was performed to establish the association between cell type proportions and AUD status as a function of smoking, age, sex, and MDD. Bioinformatics tools were also used to determine DGE related to AUD cases and to understand gene co-expression. In particular, gene co-expression analysis using weighted gene co-expression network analysis (WGCNA) revealed shared and region-specific gene networks across the NAc and PFC, further elucidating the molecular mechanisms related to AU.

Results of the study

In the NAc_LIBD and PFC_LIBD datasets, 90 and 98 differentially expressed genes (DEGs), respectively, were identified. Twelve genes were found to overlap in both datasets. No DEGs were identified among the 20,958 genes tested in the NAc_UT dataset. In the PFC_UT and PFC_NYGC datasets, 14 and 53 DEGs were recognized, respectively. These newly identified AUD-linked DEGs provided insights into AUD gene expression signatures in specific brain regions.

A total of 447 DEGs associated with AUD were identified in PFC. However, 25 genes were found to be differentially expressed in NAc and PFC that were related to AUD. The top five DEG genes identified in the meta-analysis of overlapping genes in NAc samples were ODC1, ZNF844, ARRDC3, FAM225Ai GUSBP11, and through the PFC samples were TXNIP, ODC1, HMGN2, SLC16A9i SLC16A6.

The current study identified CSPP1 as the only gene significantly related to AUD in the caudate nucleus (CN); no NAc meta-analysis genes were associated with AUD in the ventral striatum (VS) and putamen (PUT). No significant PFC meta-analysis genes were found to be associated with AUD in CN, VS, or PUT.

Gene set enrichment analysis (GSEA) for NAc and PFC meta-analyses uncovered four KEGG (Kyoto Encyclopedia of Genes and Genomes) pathways. Weighted gene coexpression network analysis between regions (WGCNA) revealed that there was no module associated with AUD. The NAc_LIBD and PFC_LIBD modules were compared, showing that 97.8% of the genes in these modules overlapped, suggesting high levels of co-expression between regions.

Therapeutic intervention for AUD

The Drug Repurposing Database tool was used to identify a potential DEG as a therapeutic target for AUD. Of particular interest, 29 pharmaceutical compounds targeting DEGs in NAc and 436 pharmaceutical compounds targeting DEGs in PFC were identified, underscoring the potential for drug repurposing to treat AUD. Of the 54 DEG genes identified in NAc, 11 genes were targeted by 29 drug compounds. In addition, 64 of the top 100 genes with AUD-associated DGE in PFC could be targeted by 436 drug compounds. Therefore, the current study uncovered potential pharmacotherapies for AUD.