Re-defininG AddiCCH3Tion: genomics and epigenomics on substance use disorders

Drug addiction is a term applied across substance use disorders (SUDs) and defined as a chronic, relapsing complex brain disease characterized by compulsive drug seeking, craving, loss of self-control, and impaired decision making (National Institute on Drug Abuse 2010). Drug addiction persists in spite of harmful consequences and cuts across all demographics. Addiction results from the interactive effects among multiple genes, acting in multiple environments, and across multiple stages of development. It is this intersection that roots SUDs, in concert with other biological processes, in genetic and epigenetic mechanisms that provide a scaffold for normal brain development, for learning and memory, and for pathophysiology. The heritability of addictions ranges from moderate to high. Twin studies have shown that genetic factors account for approximately half of the variance for addictive disorders (Kendler et al. 2008). It is thought that the remaining heritability comes from environmental and developmental aspects, including epigenetic factors that contribute to gene expression miscues of otherwise highly synchronized gene regulation. Over the last decade, genetic research on SUDs has transitioned from candidate gene and linkage methods to genome-wide association studies (GWAS) and sequencing-based approaches. Many studies have been fruitful mainly because studying the genetics of SUDs benefits from a vast knowledge of a given drug’s mechanism of action (Rutter 2006). Among these are genes encoding for the enzymes involved with drug metabolism such as alcohol and aldehyde dehydrogenases, which offer protection against alcoholism (Chen et al. 2009) and CYP2A6, which offers protections against nicotine addiction (Malaiyandi et al. 2005). Studies have also identified gene variants that modulate the response to the rewarding or aversive effects of drugs such as the nicotinic cholinergic receptor subunit gene (CHRNA5, a5) for nicotine (Fowler et al. 2011) and the l-opioid receptor gene (ORPM1) for alcohol and opioids (Arias et al. 2006). Finally genes have been identified that may help predict outcomes for naltrexone treatment for alcohol dependence, such as the A118G OPRM1 single-nucleotide polymorphism (SNP) (Enoch 2013) and for response to nicotine replacement therapy for smoking cessation, such as the D398N SNP in the a5 gene and CYP2A6 genetic polymorphisms (Chen et al. 2014a,b). As with many complex disorders, clear GWAS successes have been hard to come by. However, a striking example in the field of nicotine dependence is the a5-a3b4 nicotinic cholinergic receptor subunit gene cluster on chromosome 15q, which sparked the discovery of several loci related to smoking and its more dire health consequences, such as lung diseases and peripheral artery disease (Volkow et al. 2008a; Welter et al. 2014). While found initially via a hybrid candidate gene-GWAS design (Saccone et al. 2007), the a5-a3-b4 gene cluster was replicated by GWAS and meta-analysis approaches to become one of the most replicated variants in complex disease genetics (Liu et al.2010; Thorgeirsson et al. 2010; Tobacco and Genetics Consortium 2010). This seminal finding pointed to lesser known targets for nicotine addiction (the a5-a3-b4 nicotine subunit receptors) in addition to the more common a4/b2 nicotinic receptors, which are thought to mediate nicotine reward and dependence (Picciotto et al. 1998). This GWAS finding was also crucial for identifying the importance of the medial habenula (MHb) and the interpeduncular nucleus (IPN), brain regions that express high levels of the a5 subunit containing receptors (Changeux 2010), and which form a brain circuit that plays a role in the aversive effects of nicotine. Since the MHb-IPN inhibits the activity of brain dopamine neurons involved with reward, the a5 gene pointed to the importance of the aversive effects of nicotine in the process of addiction. Indeed a5 knockout mice are much less sensitive to the aversive effects of high nicotine doses than are wild-type mice which avoid high nicotine doses (Fowler et al. 2011). Combined with the dense expression of a5 in the MHb-IPN, a picture emerged which implicates nicotinic