Polymorphisms of the serotonin transporter and receptor genes: susceptibility to substance abuse.

Serotonin (5-hydroxytryptamine [5-HT]) is an important neurotransmitter implicated in regulating substance-use disorder (SUD) acquisition, maintenance, and recovery. During the past several years, an abundance of research has begun discovering and describing specific 5-HT genetic polymorphisms associated with SUDs. Genetic variations in the 5-HT system, such as SLC6A4, HTR1B, HTR2A, HTR2C, HTR3 (HTR3A, HTR3B, HTR3C, HTR3D, and HTR3E), likely play a role contributing to SUD patient heterogeneity. The 5-HT transporter-linked polymorphic region S allele, located in SLC6A4, has now been modestly associated with alcohol dependence in two large meta-analyses. Additional 5-HT genes may also play a role but have not been extensively investigated. A limited number of SUD treatment studies have included 5-HT gene variation as moderating treatment outcomes, but the results have been equivocal. Future research on 5-HT addiction genetics should adopt whole-genome sequencing technology, utilize large study samples, and collect data from multiple ethnic groups. Together, these methods will build on the work already conducted with the aim of utilizing 5-HT genetics in SUD treatment settings.

Introduction

Substance use disorders (SUDs) have severe negative effects on addicts, their families, communities, the economy, and society. The National Survey on Drug Use and Health reported that in 2008, 14.2% of individuals 12 years of age and older had used illicit drugs during the past year.1 The Federal Bureau of Investigation shows that 12.2% of more than 14 million arrests in 2008 were for drug violations, the most common arrest crime category.2 Fifty-three percent of state and 45% of federal prisoners met Diagnostic and Statistical Manual for Mental Disorders (DSM) criteria for SUDs.3 The annual financial cost of drug use disorders and alcohol abuse in the United States is estimated to be $180 billion and $184 billion, respectively.4 Addiction research focused on serotonin (5-HT) genetics has begun clarifying specific gene polymorphisms associated with specific SUDs and related behaviors. In the future, these 5-HT variants may inform individualized choices of treatment, ultimately reducing SUD-related problems for patients as well as the enormous public health burden caused by addictions.5,6

Polymorphisms of the serotonin transporter and receptor genes

Addiction susceptibility is influenced by both biological and environmental systems. Altered 5-HT transmission is thought to increase susceptibility to a wide range of SUDs.7 The 5-HT system is highly complex, composed of neural communication mediated by multiple pre- and postsynaptic 5-HT receptor subtypes as well as a 5-HT transporter system that brings 5-HT back into the presynaptic neuron.8 Genetic polymorphisms throughout each of these 5-HT-related genes collectively give rise to unique genetic architecture, which may ultimately contribute towards an individual’s addiction risk, disease trajectory, and treatment recovery (Table 1).

Table 1

Sample of serotonin (5-hydroxytryptamine) transporter (5-HTT) and receptor polymorphisms implicated in substance use disorder susceptibility

Gene	Location	Polymorphism	Region	Alleles	Sub-Saharan African	Caucasian	Asian	Regulatory mechanism
SLC6A4	17q11.2	5-HTTLPR	5′ upstream region	S/L/XL&VL	0.27/0.72/0.01	0.43/0.57/0.00	0.80/0.19/0.01	S allele →↓5-HTT transcription efficiency and ↓5-HT uptake in vitro100
		STin2	Intron 2	7,9, 10/12	0.31/0.69	0.38/0.62	0.09/0.91	Stin2.12 allele ↑transcription13
		rs25531	5′ upstream region	A/G	0.81/0.19	0.92/0.08	0.29/.0.3 1	LA allele →↑5-HTT mRNA levels17
HTRIB	6q14.3-q16.3	rs6296	5′ upstream region	C/G	0.24/0.76	0.34/0.66	0.55/0.44	GG genotype →↑binding to 5-HTIB in PFC Regulation of mRNA secondary structure?43
		rs1 1568817	5′ upstream region	G/T	0.17/0.83	0.41/0.59	0.09/0.91	G allele →↑5-HTI B transcription activity101
		rs 130058	5′ upstream region	A/T	0.98/0.02	0.78/0.22	0.94/0.06	T allele →↑expression activity47
HTR2A	13q14-q21	rs6313	Exon 1	C/T	0.62/0.38	0.53/0.47	0.49/0.51	Regulation of mRNA secondary structure or stability?102
		rs6311	5′ upstream region	C/T	0.61/0.39	0.53/0.47	0.48/0.52	TT genotype →↑5HT2A gene expression in cell lines and endogenously expressing 5HT2A103
HTR2C	Xq24-q28	rs6318	Exon 4	C/G	0.47/0.53	0.15/0.84	0.06/0.94	C allele expressed membranes →↑binding to metachlorophenylpiperazine (m-CPP) and 5-HT104
HTR3B	11q23	rs 1176744	Exon 5	G/T	0.43/0.57	0.36/0.64	0.32/0.68	G →↑ maximum response to 5-HT ↓desensitization and deactivation kinetics mean ↑channel open time83
		rs3758987	5′ upstream region	A/G	0.54/0.46	0.75/0.25	0.85/0.15	Regulation of gene expression?81
		rs3782025	Intron 6	C/T	0.41/0.59	0.46/0.54	0.21/0.79	Located in highly conserved region82

Abbreviations: 5-HT, serotonin transporter; mRNA, messenger ribonucleic acid; PFC, prefrontal cortex

SLC6A4 and substance use disorders

The 5-HT transporter (5-HTT) protein plays a central role in the 5-HT system by regulating the reuptake of 5-HT following synaptic release. Multiple drugs of abuse have been shown to interact functionally with 5-HTT.9,10 The 5-HTT is encoded by a single gene (SLC6A4) on chromosome 17q11.1–17q12 and contains two well-studied polymorphisms associated with susceptibility to SUDs. A common promotor polymorphism (5-HTTLPR)11 with two common alleles, a long allele with 16 repeats (L) and a short allele with 14 repeats (S), as well as other ethnic-specific variants that occur at much lower frequency, modulates 5-HTT expression with the S allele associated with reduced SERT transcriptional activity in vitro.11 STin2 VNTR, a second well-studied polymorphism of SLC6A4, consists of a17-bp variable number of tandem repeats with four different alleles that correspond to the number of tandem repeats (12, 10, 9, or 7), with the 12 and 10 most frequently observed.12 The STin2.12 allele has been reported to be a transcriptional enhancer.11,13–15 Given the importance of SLC6A4 in the 5-HT system, these polymorphisms, most of which focus on 5-HTTLPR, have been investigated as related to SUDs.

Alcohol

Feinn et al16 performed a systematic investigation of 17 published studies (including 3489 alcoholics and 2325 controls, primarily of European ancestry) and found that the frequency of the S allele at 5-HTTLPR was significantly associated with alcohol dependence (AD). The effect size was increased in individuals with a comorbid psychiatric condition, early onset or more severe AD subtype. Subsequent to the Feinn et al16 report, Hu et al (2006) described an A/G (rs25531) single-nucleotide polymorphism (SNP) within the 5-HTTLPR L rendering it triallelic (LA, LG, S). The LA is associated with high levels of 5-HTT mRNA transcription in vitro, whereas the LG is more similar to S allele, with low levels of 5-HTT mRNA.17 The most recent studies investigating 5-HTTLPR and its relationship to SUDs, mainly AD, have typically analyzed 5-HTTLPR as triallelic.18 Additionally, larger study samples, enhanced population specificity, and inclusion of negative life events such as early childhood trauma continue to clarify the relationship between 5-HTTLPR and AD.18–21 Because publications on the link between 5-HTTLPR and AD had almost doubled since the Feinn et al report, McHugh et al (2010) conducted a follow-up meta-analysis of 22 studies that included 8050 participants. A significant association between AD diagnosis and the presence of at least one S allele was found with slightly stronger results for participants who had two copies of the S allele.22 Relatively fewer studies have investigated the association between STin2 and AD and the research that has been conducted has produced equivocal results.23,24

Other drugs of abuse

SLC6A4 polymorphisms have been studied as related to nicotine dependence (ND), cocaine dependence (CD), and opioid dependence (OD), but altogether these studies are relatively fewer than on AD. Of the studies investigating the association between 5-HTTLPR and nicotine, some have shown the association with the L allele,25–28 others have shown the effect with the S allele,29–31 and several more have not observed any effects at all.18,32,33 The association between SLC6A4 (5-HTTLPR and STin2) and OD or CD are also mixed.24,34–37 Until more research is conducted, the relationship between these drug use disorders and SLC6A4 remains unclear.

HTR1B and substance use disorders

The 5-HT1B receptors are widely distributed throughout the brain. The 5-HT1B heteroreceptors are found mainly on gamma-aminobutyric acid (GABA) terminals projecting from the nucleus accumbens to the ventral tegmental area, pathways thought to have importance for drug reward.38 The human HTR1B gene, located at 6q14.3–q16.3, is intronless and encodes a 390-amino-acid polypeptide.39 While several HTR1B gene variants have been investigated as related to SUDs, rs6296 (G861C, Val3Val), a synonymous SNP in exon 1, is the best-studied gene variant.40 The function of this variant is unknown, but it may influence the secondary structure of RNA.41

Lappalainen et al (1998) examined the role of HTR1B rs6296 in vulnerability to AD with antisocial personality disorder (ASPD) and intermittent explosive disorder.42 The frequency of rs6296 C allele was observed to be elevated in Finns with antisocial alcoholism. In an American Indian replication sample, HTR1B rs6296 showed significant linkage to antisocial AD. Shortly following this work, Huang et al (1999) studied 5-HT1B receptor binding in a sample of patients with heterogeneous psychopathology, some of whom had a history of SUDs.43 Though the results did not support an association between the rs6296 and AD, they did observe that the C allele of rs6296 had lower 5-HT1B receptor binding as measured by Bmax values in postmortem human brain. Subsequent reports have attempted to clarify this signal as related to AD, with equivocal results.41,44,45 Kranzler et al (2002) studied linkage disequilibrium (LD) across three HTR1B polymorphic sites (rs11568817 [G-261T], rs6296, rs6298 [C129T]) and found no evidence for allelic association with AD and SUDs in general, alone or with a comorbid ASPD diagnosis.46 Sun et al (2002) sequenced HTR1B in a Taiwanese Han sample of individuals with AD and controls.47 Among several variants identified, rs130058 (A-161T) T allele was significantly higher in AD cases than in controls and was demonstrated, in vitro, to have higher expression than the A allele. Cao et al (2011) also found that the T allele of rs130058 was associated with AD as well as a younger age of onset of AD in a Chinese Han sample.48 Recently, larger studies have reported additional SNPs in HTR1B as related to AD. More work should be conducted to replicate these signals and clarify how these variants relate to the better-characterized SNPs.49,50 Interestingly, a functional HTR1B SNP (rs13212041) located in a microRNA binding site has been described and reported to account for a larger proportion of variation in self-reported anger and hostility compared with well-known HTR1B SNPs described above.51 To our knowledge, rs13212041 has not been extensively studied as related to AD and warrants further investigation.

Some workers have investigated whether HTR1B genetics is related to SUDs other than AD. Cigler et al (2001) studied HTR1B gene variants in 210 individuals with a primary diagnosis (DSM-IV criteria) of CD or abuse, AD or abuse, and controls. Of the seven HTR1B polymorphisms studied, including rs6296, there was no association found.52 Israeli female college students who carried two copies of the G allele at rs6296 were observed to have higher levels of ND.53 Ujike et al (2011) conducted a small case-control study of methamphetamine dependence in a Japanese sample. No association between HTR1B SNPs (rs130058, rs1228814, and rs1228814) and methamphetamine dependence was observed.54

HTR2A and substance use disorders

The 5-HT2A receptor is a ubiquitous G-protein-coupled receptor that is distributed widely throughout the central nervous system in mammals.55,56 5-HT2A receptors have been shown to modulate dopamine neurotransmission,57,58 making it a target for substance use susceptibility. The HTR2A gene is located in the chromosomal region 13q14–q21. Though the HTR2A gene is highly polymorphic, the majority of addiction genetic research has focused on two SNPs – rs6311 (A-1438G) and rs6313 (T102C). These two SNPs are reported to be in linkage disequilibrium. The rs6311 SNP is located upstream of two alternative promoters for HTR2A and may affect the promoter activity and alter the expression of HTR2A in the brain.59 The rs6313 SNP is in exon 1 of the HTR2A and codes the 34th amino acid as serine.

A small positive association was reported between the G allele of rs6311 in Japanese alcoholics with inactive ALDH2 compared with control subjects.60 A second study investigating rs6311 in a sample of individuals with AD, OD, and healthy controls observed that individuals with AD, compared with OD, were significantly enriched with the G allele, but only when they carried the 5-HTTLPR L allele.24 SNP rs6313 was not observed to moderate response to alcohol in a small sample comparing low- and high-level alcohol responders,61 nor was it found to be associated with AD patients compared to nonconsanguineous relatives.62

The influence of 5-HTR2A gene variation has also been studied in several other SUDs. The HTR2A rs6311 A allele was associated with cigarette smokers compared with nonsmoking controls in a sample of AD European-derived Brazilians.63HTR2A rs6313 was unrelated to smoking phenotypes in a large sample of young adults that included smoking status and nicotine level.64 Studies conducted thus far in OD have not observed an association with HTR2A,24,65 and studies linking this gene with CD have not been thoroughly investigated. Interestingly, rs6311 was recently considered as a potential moderator of medication overuse headache, a chronic headache caused by excessive use of medication taken for symptomatic headache. Though the results of the study were mainly negative, a second HTR2A SNP, rs6305 (C516T), was related to monthly drug consumption in medication overuse headache patients and possibly drug-seeking behavior.66

HTR2C and substance use disorders

5-HT2C receptors are densely expressed throughout multiple brain regions, and have been shown to modulate discriminative stimulus effects of cocaine.67–69 The HTR2C gene is located at the Xq24–q28 chromosomal band.70 Limited research supports the role of HTR2C sequence variation and SUDs. Lappalainen et al (1995) discovered rs6318 (Cys23Ser) encoding a missense mutation at codon 23 of the Cys-to-Ser substitution in a Caucasian population.71 Interest in rs6318 developed following the observation that the Ser substitution was associated with 3-methoxy-4-hydroxyphenylglycol (MHPG), a metabolite of norepinephrine degradation, both in AD violent offenders and population controls. The same study was unable to detect any association between rs6318 and AD or alcohol abuse.72 Subsequently, the association between rs6318 and MHPG was not replicated,73 and several other studies failed to find a relationship between rs6318 and AD.61,74 Mottagui-Tabar et al (2004) reported that none of the four HTR2C promoter SNPs investigated nor the promoter microsatellite were shown to be associated with AD.75

HTR3 and substance use disorders

The 5-HT3 receptor is the only ligand-gated ion channel in the 5-HT receptor family and may play a role in reward by modulating dopamine release in the mesolimbic pathway.76,77 GABA(A) and glycine ligand-gated ion channel receptors, sharing structural and functional homology to 5-HT3 receptors, are targets for the acute and chronic effects of alcohol.78HTR3 genes reside in close proximity on chromosome 11q23 (HTR3A and HTR3B) and 3q27 (HTR3C, HTR3D, and HTR3E) in the human genome.79,80 Relatively few studies have investigated the association between HTR3 genetics and SUDs. Levran et al (2008) studied 1350 variants in 130 candidate genes in patients with a past history of severe HD and compared them with healthy controls with no history of drug abuse. The C allele of rs3758987 in HTR3B gave a nominally significant association with OD. This SNP is located at the gene regulatory region and may impact gene expression.81 In a multisite study, the A allele of an intronic HTR3B SNP rs3782025 was associated in individuals with both AD and ASPD in a Finnish sample, but not in Bethesda and Plains Indian samples.82 A recently published study found that an HTR3B missense polymorphism rs1176744 (Tyr129Ser) predicted AD and was also related to AD and SD together.83 The Ser129 variant of rs1176744 substitution that was associated with AD has been shown to expand maximum response to 5-HT, lower desensitization and deactivation kinetics, and increase mean channel open time.84,85 The rs1176744 has also been shown to be in strong linkage disequilibrium with rs3782025.

5-HT genetics and substance use disorder treatment

5-HT genetics are not currently part of any approved and empirically based addiction treatment algorithm. However, optimism remains about potential clinical use of 5-HT genetics, as well as genes from other relevant systems (eg, dopamine, GABA, norepinephrine, opioid, etc) to treat SUDs. Embracing 5-HT genetic variations could impact treatment in three key ways. First, improved characterization of 5-HT genetics could lead towards personally optimizing addiction medications, especially drugs targeting the 5-HT system. This genetic information could be useful in predicting the therapeutic efficacy of certain medications as well as their side-effect profile, allowing for more precise treatment. Second, enhanced characterization of 5-HT variants, along with other gene systems, will lead towards more accurate SUD diagnosis. For example, 5-HT variants could predict executive cognitive functioning or externalizing traits that may influence relapse proneness via general neurobehavioral mechanisms, thus could have significant treatment moderating effects.86 Third, enhanced understanding of 5-HT gene variation could lead towards earlier identification and intervention, potentially resulting in fewer negative consequences to the individual at risk and society.

Animal studies and human clinical laboratory paradigms suggest that ondansetron, a 5-HT3 antagonist, has potential clinical utility for treating AD, but its efficacy varies widely. Research has investigated 5-HT genetics to better clarify AD ondansetron responders versus nonresponders. A randomized controlled treatment study indicated that individuals with AD treated with ondansetron and who were also homozygous for 5-HTTLPR L allele had fewer numbers of drinks per drinking day and a higher percentage of days of total abstinence. This pharmacogenetic effect was enhanced by including rs1042173, an SNP in the 3′ untranslated region of the SLC6A4 transcript.87–90 Kenna et al (2009) also observed that ondansetron improved drinking outcomes in nontreatment-seeking AD subjects carrying the L/L genotype compared with those with the L/L genotype receiving sertraline, a selective serotonin reuptake inhibitor (SSRI), who slightly increased their drinking.91 Kranzler et al (2012) conducted a 12-week double-blind randomized trial of sertraline for AD and genotyped individuals for 5-HTTLPR. In drinkers with late-onset AD, those taking sertraline, who also had the L allele, had better drinking outcomes (fewer drinking and heavy drinking days). Conversely, those with early onset AD and who carried the L allele had better drinking outcomes when treated with placebo.92 A recent study in patients with comorbid major depression and AD treated with escitalopram, an SSRI, indicated no 5-HTTLPR genotype effect on the Alcohol Use Disorders Identification Test as compared with memantine, a nonserotonergically acting medication.93 Wong et al (2008) conducted a small study to determine whether 5-HTTLPR and 5HT2C rs6318 moderated the influence of citalopram, an SSRI, on adrenocorticotrophic hormone levels in males with AD and controls. Both AD and control groups with Ser23 alleles had significantly higher adrenocorticotrophic hormone responses during placebo administration and attenuated the responses under citalopram administration.94

Mannelli et al (2005) did not observe a 5-HTTLPR-moderating effect of cocaine use in a sample of African-American CD patients during a behaviorally oriented outpatient treatment for CD.35 To date, 5-HT genetic moderators of OD treatment studies have not been conducted. However two studies, one investigating postoperative analgesic requirements after major abdominal surgery and the second looking at analgesic response to remifentanil, a potent synthetic opioid analgesic drug, provide weak evidence for investigating 5-HT genetics in future OD treatment research.95,96 Despite the apparent importance of 5-HT neurotransmission to smoking cessation, three studies investigating the influence of 5-HTTLPR on the outcome of NRT found no association between genotype and abstinence at the end of NRT treatment.97–99 David et al (2008) also studied other 5-HT gene variants (5-HTTLPR, rs1799913, rs6295) that were not associated with abstinence following NRT either.99

Conclusion

Given that the 5-HT system is modulated by all of the major classes of drugs of abuse, better characterization of 5-HT gene variation would help increase the potential to identify patients with SUD earlier and perhaps prior to illness onset, develop more refined diagnostics, and improve treatment outcomes. However, understanding how the relationship between individual 5-HT genetic architecture relates to SUD remains in its infancy, and several of the markers described here only equivocally associate with SUDs. Many of the studies described were small (<500), focused on only a few 5-HT SNPs, and were conducted in primarily Caucasian samples. Any single 5-HT SNP marker accounts for a relatively small portion of the variation in SUDs. Based on this literature review, 5-HTTLPR S allele appears to exhibit the strongest association with SUD, specifically AD. However 5-HTTLPR is also the best-studied 5-HT polymorphism. Perhaps other 5-HT polymorphisms contribute equally or even more to SUD than 5-HTTLPR, but have not been as thoroughly investigated. The studies reviewed above provide a foundational link between 5-HT genetics and SUDs. The next step towards translating SUD and 5-HT genetic research into true clinical applications will benefit from embracing whole-genome sequencing to analyze the majority of 5-HT genetic variation in a single project, capitalizing on large publicly available data sets to significantly increase the sample size of any given study, and examining samples drawn from genetically distinct ethnic ancestries. These enhanced study methods in concert with work described in this review provide hope that in the future 5-HT genetics, as well as other variants from other gene systems, will be used to better understand better SUD presentation variability and disease course while increasing treatment response.