Expression studies of naturally occurring human dopamine transporter variants identifies a novel state of transporter inactivation associated with Val382Ala

Abstract

Multiple, rare, human dopamine (DA) transporter (hDAT, SLC6A3) coding variants have been described, though to date they have been incompletely characterized. Here we present studies analyzing the function and regulation of five naturally occurring coding variants, V55A, R237Q, V382A, A559V and E602G, expressed in COS-7 and SH-SY5Y cells. All variants, except V382A, exhibited levels of surface protein expression and DA transport activity comparable to hDAT. V382A, divergent at the most highly conserved residue among reported hDAT variants, exhibited significantly diminished surface expression, likely derived from inefficient plasma membrane delivery. Moreover, a greater expression of V382A protein was required to achieve comparable levels of transport to hDAT, consistent with a loss of transport function. V382A displayed a decrease in sensitivity to phorbol ester (PMA)-induced internalization, as well as an altered substrate selectivity for DA versus norepinephrine (NE). Analysis of PMA-induced V382A internalization revealed a trafficking-independent action of PMA, consistent with the existence of a surface-localized, transport-inactive state.

Introduction

An important mechanism for inactivation of dopamine (DA) following release is reuptake via the dopamine transporter (DAT). Human DAT (hDAT, SLC6A3) is a member of the Na⁺/Cl⁻-coupled cotransporter gene family that also includes the transporters for norepinephrine (NE) and serotonin (5HT) (Chen et al., 2004). The human dopamine transporter (hDAT) gene contains 15 exons, encoding a 620-amino acid protein (Giros et al., 1992). The hDAT protein is predicted to have 12 transmembrane domains, a large extracellular second loop with multiple glycosylation sites, and multiple consensus sites for phosphorylation, although the exact sites of phosphorylation remain to be determined (Giros et al., 1992, Granas et al., 2003, Li et al., 2004, Vaughan, 2004). The hDAT can assemble as homomultimers (Hastrup et al., 2001, Torres et al., 2003), as well as interact with several other proteins including PICK-1, Hic-5, PP2Ac, and syntaxin 1 (Bauman et al., 2000, Torres et al., 2001, Carneiro et al., 2002, Lee et al., 2004). The activity of hDAT is dynamic and can be modulated, usually by increasing or decreasing hDAT's presence at the plasma membrane through multiple second messengers and protein kinase-linked pathways (Melikian and Buckley, 1999, Saunders et al., 2000, Carvelli et al., 2002, Granas et al., 2003, Loder and Melikian, 2003, Moron et al., 2003). Additionally, hDAT activity can be regulated by trafficking-independent mechanisms including modulation of DAT-mediated currents and DA efflux (Sonders et al., 1997, Khoshbouei et al., 2004).

The physiological importance of DAT is evident by work from DAT knock-out mice. DAT knock-out mice display a hyperdopaminergic phenotype including a greatly decreased DA clearance rate, downregulation of D1 and D2 DA receptors with an almost compete lack of autoreceptor activity, hyperlocomotion with an inability to habituate, and cognitive deficits (Giros et al., 1996, Gainetdinov et al., 1999, Jones et al., 1999, Spielewoy et al., 2000). The hyperlocomotive phenotype of DAT knock-out mice is sensitive to treatment with the psychostimulants, amphetamine and methylphenidate, leading to the idea that the DAT knock-out mouse may serve as a model of attention deficit hyperactivity disorder (ADHD) (Gainetdinov et al., 1999). The link between DAT and ADHD is not limited to animal studies, as both genetic and imaging studies have been conducted in human subjects with ADHD to address whether DAT dysfunction might contribute to ADHD. Association studies using a variable number tandem repeat in the 3′untranslated region (3′VNTR) of hDAT have found that the 480 base pair 10-repeat allele is associated with ADHD in the majority of cases (Cook et al., 1995, Gill et al., 1997, Waldman et al., 1998, Daly et al., 1999, Palmer et al., 1999, Holmes et al., 2000, Barr et al., 2001, Curran et al., 2001, Todd et al., 2001, Chen et al., 2003). Additionally, imaging studies using a radiolabeled antagonist of DAT have shown that DAT binding is higher in subjects with ADHD and this elevation is decreased to control levels after subjects are treated with methylphenidate (Dougherty et al., 1999, Dresel et al., 2000, Krause et al., 2000, Cheon et al., 2003).

Given suggestions of hDAT dysfunction in human disease, several screens of the hDAT coding region for functional polymorphisms have been conducted. Four studies to date have examined the coding region and splice junctions for polymorphisms in hDAT (Cargill et al., 1999, Grunhage et al., 2000, Vandenbergh et al., 2000, Mazei-Robison et al., 2005). Five nonsynonymous polymorphisms (V55A, R237Q, V382A, A559V, and E602G) were identified between these four studies. The R237Q mutation was identified in a large screen for coding single nucleotide polymorphisms (SNPs) that examined 106 genes in an average of 57 subjects (Cargill et al., 1999). The V55A and V382A mutations were identified in a screen for hDAT SNPs in 150 control subjects, 109 subjects diagnosed with Tourette's syndrome, 64 alcoholics, and 15 subjects diagnosed with ADHD (Vandenbergh et al., 2000) and were recently characterized in vitro (Lin and Uhl, 2003). The A559V and E602G mutations were identified in an hDAT screen of 70 individuals with bipolar disorder, with no mutations identified in 46 control subjects (Grunhage et al., 2000). Additionally, we recently re-identified the A559V mutation in affected siblings in a screen of 70 children with ADHD (Mazei-Robison et al., 2005).

In parallel with a search for novel hDAT gene variants in candidate disorders, we felt it was important to systematically characterize the impact of all of the known nonsynonymous variants using multiple in vitro expression systems. Using transiently transfected COS-7 and SH-SY5Y cells, we found that only V382A significantly affects basal transport activity. This effect derives from both a decrease in cell surface expression and from a loss of catalytic efficiency. Consistent with the altered efficiency of V382A, we also obtained evidence for an altered DA:NE selectivity, suggesting that V382A displays an altered conformation at the cell surface. This alteration of V382A activity was further amplified by the observation that the variant is not as readily internalized in response to treatment with the phorbol ester (phorbol-12-myristate-13-acetate, PMA). PMA can still elicit a robust downregulation of V382A transport activity but there is not a concomitant increase in internalization, suggesting that the presence of V382A may uncover an unrecognized mode of hDAT modulation whereby surface-localized transporters shift to an inactive state prior to internalization. We discuss our findings with respect to the emerging evidence that neurotransmitter transporter variants can impose both basal and regulatory alterations and that transporters appear to be modulated through both trafficking-dependent and -independent mechanisms.