The dopamine transporter gene SLC6A3: multidisease risks

Abstract

The human dopamine transporter gene SLC6A3 has been consistently implicated in several neuropsychiatric diseases but the disease mechanism remains elusive. In this risk synthesis, we have concluded that SLC6A3 represents an increasingly recognized risk with a growing number of familial mutants associated with neuropsychiatric and neurological disorders. At least five loci were related to common and severe diseases including alcohol use disorder (high activity variant), attention-deficit/hyperactivity disorder (low activity variant), autism (familial proteins with mutated networking) and movement disorders (both regulatory variants and familial mutations). Association signals depended on genetic markers used as well as ethnicity examined. Strong haplotype selection and gene-wide epistases support multimarker assessment of functional variations and phenotype associations. Inclusion of its promoter region's functional markers such as DNPi (rs67175440) and 5'VNTR (rs70957367) may help delineate condensate-based risk action, testing a locus-pathway-phenotype hypothesis for one gene-multidisease etiology.

Fig. 1. Study design.

Eight channel-synthesis of risk knowledge for functionally polymorphic and vulnerable SLC6A3, allowing proposing of a multidisease hypothesis. Mustard oval, reviewed data; orange (1/molecular modeling, 3, 4, 5, 7, 8), new data whose collections are described in the same order in Methods section; blue arrow, mechanism; light blue arrow, functional approach; IDRs of TFs, intrinsically disordered regions of transcription factors.

Fig. 2. Functional syntheses for association signals.

a Molecular modeling of familial hDAT mutants for structural interpretation. (left) Two-dimensional distribution of mutations. Each gray cylinder represents a transmembrane (TM) domain; upside, synaptic space; downside, cytosol of DA neuron. (right) Three-dimensional structural model of mutated residues with TM helices represented by gray cylinders. Mutated residues are represented as licorice sticks and colored using the same convention as in (left panel): DTDS in orange, autism in blue, ADHD in red, BP in green and the combined mutants for autism, BP and ADHD in magenta and labeled. The substrate/inhibitor site that includes both S1 and S2 sites is represented by the docked inhibitor represented in space filling model and colored yellow (these represent new information - see Supplementary text for Fig. 2a in the Supplementary Information). b Associated (underlined), functional (>) or other complex (orange symbol) markers in regulated SLC6A3. SLC6A3 has 15 exons (14 introns), with the start codon in Exon 2 and the stop codon in Exon 15 so that its 1863 bp-coding region ends with the first 23 bp of the 2 kb long Exon 15 (based on GenBank Accession # NM_001044). Indicated are functional markers in red and other unique complex markers in orange: 3’VNTR and rs27072 in 3’UTR, Int8VNTR, rs64049 and rs393795 in Intron 4, DNPi and SSLP (▲) in Intron 1, rs2975226 in core promoter, rs1478435, 5’VNTR (@) and -14k-VNTR (♥) in distal promoter regions. SNPs are indicated with ↓. Green, SLC6A3 gene structure (vertical bars are 15 exons in the opposite strand of the chromosome); >, variant activity greater than; -, unknown activity for two alleles; /, a genotype associated; italics, associated diseases or response to medication (℞) (BP bipolar, METH methamphetamine use disorder, LID levodopa-induced dyskinesia, SCZ schizophrenia); bold, meta-analysis result and font size indicates significance or sample size used where same color matches risk allele with phenotype;?, only postmortem mRNA correlational data; gray: recombination rate (new data) obtained from combining all 25 1KGP populations, below with the chr5 position in kb. Paucity, lack of association efforts in promoter regions (no familial mutants are included). Only association studies with meta-analysis statistical significance or reproduceable results are summarized here.

Fig. 3. Genotype-correlation of hDAT availability (Bmax/KD in Y axis) in healthy ventral striatum (left), putamen (center) or caudate (right).

Insert, postmortem hDAT mRNA levels in isolated nigral DA neurons from twenty independent healthy subjects; upper, DNPi; middle, 5’VNTR; bottom, haplotype of DNPi and 5’VNTR; multivariate results for three brain regions are given in F values and only significant P values are shown (N = 62). Between-genotype differences were not statistically significant after Bonferroni corrections. After excluding women (N = 3), the effects remained for men. These are all new data.

Fig. 4. Genetic evidence for functional variants throughout SLC6A3, based on haplotype selection of 18 kb promoter (a) or 70 kb whole gene (b) and on case-control intragenic epistases for PD and SUDs both in Caucasians (c).

In first two panels of phylogenic trees, labeling of the 26 1KGP populations are explained as follows. East Asians all in red: closed circle for Chinese Han Beijing (CHB); open square, Japanese (JPT); closed square, Chinese Han South (CHS); closed diamond, Chinese Dai in Xishuangbanna (CDX); open triangle, Kinh in Chi Minh City, Vietnam (KHV). European ancestry all in blue: closed circle, Utah residents with Northern and Western ancestry (CEU), open diamond, Toscani in Italia (TSI); closed triangle, Finnish in Finland (FIN), closed square, British in England and Scotland (GBR); open square, Iberian population in Spain (IBS). African ancestry all in black: closed circle, Yoruba in Ibadan, Nigeria (YRI); closed diamond, Luhya in Webuye, Kenya (LWK); closed triangle, Gambian in Western divisions of the Gambia (GWD); inverse closed triangle, Mende in Sierra Leone (MSL); closed square, Esan in Nigeria (ESN); open circle, Americans of African Ancestry in SW USA (ASW); open diamond, African Caribbeans in Barbados (ACB). Admixed Americans all in green: closed circle, Mexican ancestry from Los Angles USA (MXL); closed diamond, Puerto Ricans in Puerto Rico (PUR); open circle, Colombians in Medellin of Colombia (CLM); open triangle, Peruvians in Lima of Peru (PEL). South Asian all in gray: closed circle, Gujarati Indian in Huston of Texas USA (GIH); closed diamond, Punjabi in Lahore of Pakistan (PJL); closed square, Bengali of Bangladesh (BEB); open square, Sri Lankan Tamil from the United Kingdom (STU); and closed triangle, Indian Telugu from the United Kingdom (ITU). In panel (c), 6572 unrelated subjects in three cohorts were used for PD (above the gene schematic related to Fig. 2b) and 5843 in another three cohorts, for SUDs (below the gene schematic). Shown are intragenic epistases selected largely for those involving Fig. 2b-mentioned markers where yellow indicates suggestive significance and red, for statistical significance. Unlabeled SNPs are rs11564770 (at 1398806) and rs11564772 (at 1398007) in Intron 14 or last intron, and rs250686 (at 1425159) in Intron 4 for PD; rs11564757 (160 bp downstream of DNPi indicated by asterisk) in Intron 1 and rs2975292 (at 1419932) in Intron 6 for SUDs. rs150052082, in gray, was a new SNP which interacted with multiple loci in SUDs, see Supplementary Table 1 for details. These are all new data.

Fig. 5. Multidisease hypothesis for variant- and pathway-related condensate in SLC6A3 transcription.

Gray curve, chromosome DNA harboring SLC6A3; orange rectangle, polymorphic sites (unlabeled for Intron 4); gray rectangle, cis-acting element; red oval, repressor; green oval, activator; gray oval, unknown TF as indicated by “?”; light blue dot, other condensation proteins such as Mediator; large blue circle, dynamic condensate; gray arrow, transcription start with RNA polymerase II (RPII) complex; TFs, other known transcription factors such as NURR1, GMEB1, PITX3, FOXA2, LMX1A and SP1/3. Yellow double-arrows, different disease-related and TF-mediated pathways. HEY1 has not been shown with allelic binding yet; majority of the proteins each have multiple IDRs so that this schematic has neither molecular nor dimensional accuracy.

Fig. 6. SLC6A3 genetics x age-dependent expression of phenotypes.

Horizontal blue bar, lifespan from newborn at left to elderly at right; associated phenotypes are indicated on the top where red, for psychiatric disorders and black, for movement disorders at the two vulnerable ends of lifespan; associated genetic effects are indicated below the bar; x, interaction.