“TO BE OR NOT TO BE” GWAS Ends the Controversy about the DRD2 Gene as a Determinant of Reward Deficiency Syndrome (RDS)

”To be or not to be?—that is the question” is the first line of arguably the most famous speech in Shakespeare’s Hamlet.

Early Evidence and Controversy

In 1990, Kenneth Blum and Ernest Noble set the world on its heels with the publication of their seminal research, the discovery that the A1 allele of the dopamine D2 receptor gene correctly classified 77% of subjects known to have severe Alcohol Use Disorder (AUD) and concurrently was not present in 72% of subjects without AUD.¹ Positive associations with other addictions^2–8 were also found. Blum also pointed out that the DRD2 A1 was reward linked.¹ There is confirming evidence⁹ throughout the literature that supports the original objective research by Noble and Blum et al,¹⁰ revealing that independent of alcoholism, carriers of the DRD2 A1 allele progressively reduced (the number of binding sites) Bmax in subjects with A2/A2, A1/A2, and A1/A1 alleles, with subjects with A2/A2 having the highest mean values, and subjects with A1/A1, the lowest (up to 40% reduction). These values were confirmed by others.^6,10

While NIDA scientists excitedly embraced this novel finding by Blum et al,¹ the opposite was true for NIAAA scientists. An editorial by NIAAA scientists,¹¹ and others accompanying the original JAMA report,¹ while reasonable at the time, initiated the long-standing controversy over the role of the DRD2 TaqA1 allele and its putative association with alcoholism.

The then director of the National Institute of Mental Health (NIMH), Elliot Gershon, published an article in Science Magazine showing a shattered glass castle, suggesting that Blum and Noble live in a glass castle.¹² Along these lines, in 1990, Bolos and others from the section on Genetic Studies, NIAAA did not support a widespread or consistent association between the D2 receptor gene and alcoholism.¹³ We attribute this to poorly screened controls.

Despite positive reports linking the DRD2 gene Taq A1 allele to some RDS behaviors, the controversy gained steam with negative thinking by not only scientists¹⁴ but science reporters.¹⁵ Gelernter’s group¹⁶ from Yale also did not find support for the DRD2 TaqA1 allele and alcoholism. Careful review and reappraisal^2,12 revealed two important caveats which help explain the lack of association.¹ First, the control population was derived from a French Tourette’s Cohort.² Secondly, subjects with elevated SGOT liver enzyme levels were excluded (leaving not very severe alcoholics). The poor screening of controls (or, the inclusion of hidden RDS behaviors like obesity, ADHD, PTSD, and gambling) may have produced a series of spurious results in gene-based investigations of the role of DRD2 gene polymorphisms and neuropsychiatric disorders. These conflicting results called for “super controls” in the research.¹⁷ The concept of super controls has emerged especially in the field of addiction psychiatry and is an important concept that has not properly been addressed. If one considers the old Plomin idea that one-gene-one-disease or OGOD, elimination of just this one genetic polymorphism in the controls is a simple research task. However, this is not the case when we consider complex polygenic variants with a small effect size for each individual gene polymorphism, as, for example, with an array of addictive behaviors in studies dealing with both substance and non-substance behavioral addictions. In this scenario, there could be multi-genetic loci and accompanying sometimes unobvious when screening for controls, a few RDS-type of behaviors, that must be eliminated from so-called healthy controls. If this is not accomplished the entire experiment may be flawed providing spurious and false results. To be clear, it is like analyzing a disease with a disease instead of controls. In fact, Chen et al¹⁷ revealed that assessment of the DRD2 A1 allele in unscreened or poorly screened controls was found to be 33% compared to elimination of every RDS behavior in the probands and their families, eliminating the RDS disorder in the controls resulting in only 3.3% of the super-controls.

GWAS Ends the Controversy

In 2004, Neville et al found that the DRD2 Taq1A RFLP is a single nucleotide polymorphism (SNP) that causes an amino acid substitution within the 11th ankyrin repeat of ANKK1 (p.Glu713Lys) helped to explain the previously described associations between the DRD2 Taq1A RFLP and neuropsychiatric disorders such as addiction. The polymorphic pattern of this receptor gene suggests that susceptibility to at least one form of alcoholism is conferred by a gene located on the q22-q23 region of chromosome 11.¹⁸ Blum et al also pointed out that the DRD2 A1 allele was not specific for alcoholism but was reward linked.¹

In 2008, Gelernter’s group¹⁹ reported that for both AUD and drug dependence, the ANKK1 exon 8 to DRD2; C957T was significantly associated (p = 0.0028) in both samples. Other naysayers like David Goldman became less abrasive, and by 2019,²⁰ his group reported applying a meta-analysis involving 62 studies of DRD2 and AUD with 16,294 participants and found the rs1800497 SNP associated with AUD (odds ratio, 1.23; 95% CI, 1.14–1.31; P < 0.001). They correctly pointed out that the association was attributable to spuriously low allele frequencies in controls in positive studies.²⁰ Although this appears to be a negative comment, Blum et al have consistently argued that fewer RDS behaviors in controls (thus the lower frequency of DRD2 A1 and other alleles that induce functional hypodopaminergia) are mandatory, expected, and desirable for the accurate candidate and GWAS genetic investigations.²¹ Blum’s concept of RDS characterizes a group of behaviors associated with the relative failure of the dopaminergic system. RDS behaviors are brain reward mechanisms typified by dopaminergic dysfunction, an acute excess or chronic deficit of dopamine release in the brain reward circuitry.²² The reward deficiency behaviors include drug and non-drug addictive, compulsive, and impulsive behaviors.^23,24

In 2014, Blum’s group developed the genetic addiction risk severity (GARS) test of eleven SNPs from ten genes and neurotransmitter pathways hypodopaminergic antecedents to addiction psychiatry.²⁵ In 2022, 74,566 case-controls AUD were used to statistically validate the selection of the risk alleles measured by GARS and showed significance for DRD2, DRD3, DRD4, DAT1, COMT, OPRM1, and 5HTT at 5%. These alleles captured a post-risk estimate for 8% of the population’s alcoholism prevalence.^25,26 Also, over 3000 people presenting with polysubstance abuse from at least one-dozen chemical dependency and behavioral addiction clinics, including general population mixed gender and race, were genotyped, resulting in a GARS score for drug and alcohol risk at over 90% and 72%, respectively.²⁷

The candidate association of the DRD2 A1 allele initially met with significant controversy^1,13 the gene and associated polymorphisms have now been confirmed in several elaborate GWAS studies. Levey et al,²⁷ from Yale, reported on a large meta-analysis of depression that used data from 23andMe, the Million Veteran Program, the UK Biobank, the FinnGen biobank, and subjects of European ancestry (n = 1,154,267; 340,591) and African ancestry (n = 59,600; 25,843). Remarkably, transcriptome-wide association study analyses showed significant associations with the expression of NEGR1 (a dopamine regulatory gene) in the hypothalamus and DRD2 in the nucleus accumbens, among others. This extensive investigation underscored the genetic architecture of depression and provided new insight into the interrelatedness of complex psychiatric traits.

Similar work from Kimbrel et al²⁸ carried out a GWAS that identified gene polymorphic associations in pan-ancestry and ancestry with attempted suicide among veterans. They found a robust pan-ancestry signal at the DRD2 locus (p = 1.77×10-7). Moreover, they also identified and subsequently replicated the association in a large, independent international civilian cohort (p = 7.97×10-4). Of further interest, ancestry-specific genome-wide significant loci were also identified in African Americans, European-Americans, Asian Americans, and Hispanic Americans. A pathway analysis yielded an impressive list of reward gene polymorphisms with high clinical significance, including glutamatergic synapse, dopaminergic synapse, oxytocin signaling, cortisol synthesis and secretion, and circadian rhythm. Most importantly, the authors suggest that their pathway analyses suggest that many commonly impacted biological pathways could inform the development of beneficial therapeutics for suicide prevention. These genetic findings support the RDS construct.²⁸

Finally, a new study confirms the RDS construct originally suggested by Blum’s group in 1995²³ and 1996²⁴ and could eventually lead to universal therapies to treat multiple substance use disorders. To help end the three-decade controversy, Hatoum et al²⁹ observed polymorphic alleles of DNA that underlie SUD with multiple substances, including 19 SNPs associated with general addiction risk and 47 DNA variants linked to the specific drug of abuse: 32 for tobacco, nine for alcohol, five for cannabis, and one for opioids. The DRD2 gene and associated polymorphism were the top gene signal (DRD2 (P = 7.9 × 10–12)), mapped via chromatin refolding. They found that repeated exposures to addictive substances can result in adaptation in the dopamine pathway, effect tolerance, and craving. Importantly, these analyses highlight that the regulation or modulation of dopaminergic gene function, rather than just the presence of variation in dopaminergic genes, is central to general addiction liability. In addition, it is genuinely interesting that gene therapy, as first studied by Panayotis Thanos and Nora D Volkow and others, consistently revealed that increasing the DRD2 receptors via an adenoviral vector delivered into the nucleus accumbens of rats significantly reduced alcohol intake as well as cocaine self-administration.^30–32

Evidence that dopamine neurotransmission is a top candidate with significant associations in candidate and GWAS studies involving millions of people presenting with depression, suicide ideation, attempted suicide, and SUD is an incontrovertible basis for extensive studies to identify effective treatments for neuropsychiatric diseases.