Claus J Loland and Ulrik Gether

Sobriety Success

At Home Drug Withdrawal

Get Instant Access

CONTENTS

14.1 Introduction 225

14.2 Neurotransmitter Transporters Belonging to the SLC6 Family 227

14.2.1 Structures and Mechanisms of SLC6 Transporters 228

14.2.2 The Binding Sites for Na+ and Cl-: Importance in Substrate Binding and Translocation 231

14.2.3 Substrate Specificity and Binding Sites in SLC6

Neurotransmitter Transporters 232

14.3 Drugs Targeting Biogenic Amine Transporters: Specificity, Use, and Molecular Mechanisms of Action 233

14.3.1 Cocaine, Benztropine, and Other Tropane Class Inhibitors 233

14.3.2 Amphetamine and Other Nonendogenous Substrates 236

14.3.3 Antidepressants 236

14.3.4 Other Biogenic Amine Transporter Inhibitors 237

14.4 Inhibitors of Glycine and GABA Transporters: Specificity, Use, and Molecular Mechanism of Action 237

14.5 Conclusion 238

Further Readings 238

14.1 INTRODUCTION

The transport of solutes across membranes is of fundamental importance for all living organisms and is mediated via specific integral membrane proteins. The transport processes are often energetically coupled, either directly through the hydrolysis of ATP by the transport protein itself or indirectly by the use of transmembrane ion gradients that enable the transport of the substrate against its concentration gradient. A vast amount of different transport proteins are found in both prokaryotic and eukaryotic organisms, transporting everything from nutrients and metabolites to ions, drugs, proteins, toxins, and transmitter molecules. If ion channels (see Chapter 13) are excluded, major classes of transport proteins in humans encompass ATP-driven ion pumps (e.g., the ubiquitously expressed Na-K ATPase), ATP-binding cassette (ABC) transporters (e.g., the cystic fibrosis transmembrane conductance regulator and the multidrug resistance transporter p-glycoprotein), cytochrome B-like proteins, aquaporins (water transporters), and the solute carrier superfamily (SLC) (http://www.bioparadigms.org).

The immense functional heterogeneity among transporters is illustrated by the fact that the SLC gene family alone consists of nothing less than 46 different subfamilies (http://www.bioparadigms. org/slc/menu.asp). These include several types of plasma membrane transporters, such as, high-affinity glutamate transporters (SLC1), sodium-glucose cotransporters (SLC5), sodium-coupled neurotransmitter transporters (SLC6), and a variety of ion exchangers (SLC8 + 9 + 24). The SLC

gene family also includes intracellular vesicular transporters, such as the vesicular glutamate transporters (SLC17), the vesicular monoamine transporters (SLC18), and the vesicular inhibitory amino acid transporters (SLC32).

Notwithstanding the huge number of transport proteins present in the human body, relatively few of them are targets for the action of drugs. It might even be argued that transport proteins are relatively overlooked as drug targets in spite of their critical physiological functions and some real "success stories," such as, inhibitors of the gastric ATP-driven proton pump, used against peptic ulcers, and inhibitors of monoamine transporters, used against depression/anxiety disorders (see Chapter 18). In this chapter, we focus on the monoamine transporters and then on the neurotransmitter transporters belonging to the SLC6 family (also named neurotransmitter:sodium symporters or Na-/Cl-dependent transporters) (Table 14.1). Indeed, the SLC6 transporters represent important targets for several drugs including not only medicines used against depression/anxiety but also against

TABLE 14.1

SLC6 Gene Family Neurotransmitter Transporters

Endogenous

Synthetic

Therapeutic

Substrates

Substrates

Potent Inhibitors

Use/Potential

Dopamine transporter

Dopamine

Amphetamine

CFT, GBR12,909

ADHD (amphetamines),

(DAT)

Norepinephrine

MPP+

benztropine, mazindol,

Parkinsonism?

Epinephrine

RTI-55, Cocaine, Zn2+

(inhibitors)

Serotonin transporter

5-HT

p-Cl-

Citalopram, escitalopram,

Depression, anxiety,

(SERT or 5-HTT)

Amphetamine

fluoxetin, paroxetin,

OCD (inhibitors)

MDMA

sertraline, imipramine,

("ecstasy")

cocaine, RTI-55

Norepinephrine

Dopamine

Amphetamine

Nisoxetine, nortriptyline,

Depression (inhibitors)

transporter (NET

Norepinephrine

MPP+

desipramine, duloxetine,

or NAT)

Epinephrine

venlafaxine, mazindol

Glycine transporter 1

Glycine

(R)NFPS (ALX5407),

Schizophrenia?

(GlyTl)

NPTS, Org24598

Psychosis?

Dementia? (inhibitors)

Glycine transporter 2

Glycine

ALX1393, ALX1405

Anticonvulsant?

(GlyT2)

Org25543

Analgesic? (inhibitors)

GABA transporter 1

GABA

Tiagabine,

Epilepsy (tiagabine)

(GAT-1)

SKF89976A, THPO,

exo-THPO

GABA transporter 2

GABA

?

(GAT-2 equivalent

beta-Ala

to mouse GAT-3)

GABA transporter 3

GABA

SNAP5114

?

(GAT-3 equivalent

beta-Ala

to mouse GAT-4)

GABA transporter 4

GABA

EF-1502, NNC052090,

Epilepsy? (inhibitors)

(GAT-4, equivalent to

Betaine

THPO Zn2+

EF-1502

BGT1 or mouse GAT-2)

BGT1 or mouse GAT-2)

Note: The known neurotransmitter transporters belonging to the SLC6 gene family with their respective substrates, inhibitors, and potential therapeutic use.

Abbreviations: CFT, 2P-carbomethoxy-3P-(4-fluorophenyl)tropane; MDMA, 3, 4-methylenedioxymethamphetamine; MPP+, 1-methy l-4-phenylpyridinium; (R)NFPS, N-[3-(40-fluorophenyl)-3-(40-phenylphenoxy) propyl]sarcosine; NPTS, N-[3-phenyl-3-(40-(4-toluoyl) phenoxy)propyl]sarcosine; THPO, (4,5,6,7-tetrahydroisoxazolo[4,5-c] pyridine-3-ol); ADHD, attention deficit hyperactivity disorder; OCD, obsessive-compulsive disorder. Exo-THPO: 4-amino-4,5,6,7-tetrahydrobenzo[^]isoxazol-3-ol.

obesity and epilepsy as well as drugs of abuse such as cocaine, amphetamine, and "ecstasy." Of interest, high-resolution crystal structures have recently become available of bacterial homologues of SLC6 transporters opening up entirely new possibilities for understanding how these transporters operate at a molecular level and how their function can be altered by different types of drugs.

14.2 NEUROTRANSMITTER TRANSPORTERS BELONGING TO THE SLC6 FAMILY

The availability in the synaptic cleft of the neurotransmitters dopamine, serotonin, norepinephrine, glycine, and y-amino butyric acid (GABA) is tightly regulated by specific transmembrane transport proteins belonging to the SLC6 family (Figure 14.1 and Table 14.1). The transport proteins are situated either in the presynaptic membrane or on the surface of adjacent glial cells, where they mediate the rapid removal of the released neurotransmitters and thereby terminate their effect at the pre- and postsynaptic neurons. Inside the presynaptic nerve endings, specific vesicular transporters sequester the neurotransmitters into vesicles, making them ready for subsequent release into the synaptic cleft upon the arrival of the next stimulus. The plasma membrane neurotransmitter transporters serve three main purposes: first, the transport proteins increase the rate by which the released neurotransmitters are cleared from the synaptic cleft. This rapid removal of the released neurotransmitters allows for 100-fold faster termination of neurotransmission than is possible with simple diffusion. Second, reuptake may prevent diffusion of the neurotransmitters away from the

FIGURE 14.1 The role of neurotransmitter transporters in synaptic signaling. Neurotransmitters are sequestered into synaptic vesicles through vesicular monoamine transporters (VMAT1-2) belonging to the SLC18 gene family or through vesicular inhibitory amino acid transporters (VIAAT) belonging to the SLC32 gene family. Upon arrival of an axon potential, the synaptic vesicle releases its content of the neurotransmitter into the synaptic cleft by fusion of the vesicle with the plasma membrane. The neurotransmitter exerts its effects by activating ionotropic receptors, (ligand-gated ion channels), such as GABAA receptors, glycine receptors, and 5-HT3 receptors or via G-protein-coupled receptors (GPCRs) such as dopamine receptors, adrenoceptors, 5-HT receptors, and metabotropic GABAb receptors. The fast removal of the neurotransmitter from the synaptic cleft is governed by the neurotransmitter transporter belonging to the SLC6 family located on the presynaptic neuron (DAT, SERT, NET, GlyT2, GAT-1, and GAT-2) or on glia cells (GlyT-1, GAT-1, GAT-2, and GAT-3). The neurotransmitter taken up by the presynaptic neuron allows recycling with a presumed savings in synthetic cost.

SLC6 gene family members:

• Dopamine transporter (DAT)

• Serotonin transporter (SERT)

• Norepinephrine transporter (NET)

• Glycine transporters (Glyt-1, Glyt-2)

• Betaine transporter (BGT1)

• Taurine transporter (TAUT)

Creatine transporters (CT1 and CT2)

• Proline transporter (PROT)

• Cationic amino acid transporter (ATB[0+])

• Large neutral amino acid transporter (SBAT1)

• Neutral amino acid transporter (B0AT1)

• Imino acid transporter (SIT1)

• Three orphan transporters

• Prokarytotic homologues in bacteria and archaea (>200 different, e.g., LeuT from Aquifex aeolicus)

FIGURE 14.1 The role of neurotransmitter transporters in synaptic signaling. Neurotransmitters are sequestered into synaptic vesicles through vesicular monoamine transporters (VMAT1-2) belonging to the SLC18 gene family or through vesicular inhibitory amino acid transporters (VIAAT) belonging to the SLC32 gene family. Upon arrival of an axon potential, the synaptic vesicle releases its content of the neurotransmitter into the synaptic cleft by fusion of the vesicle with the plasma membrane. The neurotransmitter exerts its effects by activating ionotropic receptors, (ligand-gated ion channels), such as GABAA receptors, glycine receptors, and 5-HT3 receptors or via G-protein-coupled receptors (GPCRs) such as dopamine receptors, adrenoceptors, 5-HT receptors, and metabotropic GABAb receptors. The fast removal of the neurotransmitter from the synaptic cleft is governed by the neurotransmitter transporter belonging to the SLC6 family located on the presynaptic neuron (DAT, SERT, NET, GlyT2, GAT-1, and GAT-2) or on glia cells (GlyT-1, GAT-1, GAT-2, and GAT-3). The neurotransmitter taken up by the presynaptic neuron allows recycling with a presumed savings in synthetic cost.

synapse of their release, thereby minimizing chemical crosstalk between adjacent synapses. Third, transporters allow recycling by reuptake of transmitters into the nerve terminal with presumed savings in synthetic cost. The crucial physiological role of the neurotransmitter transporters has been cemented by gene knockout experiments. In case of, e.g., the dopamine transporter (DAT), the disruption of the transporter gene in mice revealed the unequivocal importance of this carrier in the control of locomotion, growth, lactation, and spatial cognitive function.

The SLC6 transporters include not only the transporters of neurotransmitters (Table 14.1, Figure 14.1) but also transporters of amino acids, metabolites (creatine), and osmolytes (betaine and taurine) (Figure 14.1). Moreover, a large number of homologues have been identified in archaea and bacteria. The function of the majority of these transporters is still unknown; however, a few of them have been identified as amino acid transporters, such as, e.g., the leucine transporter LeuTAa from the Aquifex aeolicus bacterium and the tyrosine transporter Tyt1 from Fusobacterium nucleatum.

At the molecular level, the SLC6 family transporters operate as Na+ dependent cotransporters that utilize the transmembrane Na+ gradient to couple the "downhill" transport of Na+ with the "uphill" transport (against a concentration gradient) of their substrate from the extracellular to the intracel-lular environment. The transport process is so efficient that, e.g., the serotonin transporter (SERT) can accumulate internal serotonin (5-HT) to concentrations 100-fold higher than the external medium when appropriate ion gradients are imposed. Most SLC6 transporters are also cotransport-ers of Cl- and, accordingly, SLC6 transporters have been referred to as the family of Na+/Cl--dependent transporters.

14.2.1 Structures and Mechanisms of SLC6 Transporters

It is generally believed that SLC6 transporters function according to an alternating access model, which suggests a transport mechanism in which, at any given time, only the substrate-binding site is accessible to either the intracellular or the extracellular side of the membrane. Thus, at all times an impermeable barrier exists between the binding site and one side of the membrane, but the barrier can change from one side of the binding site to the other, giving the site alternate access to the two aqueous compartments that the membrane separates. A prerequisite for this model is the existence of both external and internal "gates," i.e., protein domains that are capable of occluding access to the binding site of the substrate from the external and internal domains, respectively (Figure 14.2).

In the absence of high-resolution structural information, however, it was, for a long time, only possible to speculate about the molecular basis of the transport process. A major breakthrough came when the bacterial homologue, LeuTAa, which displays 20%-25% sequence identity to its mammalian counterparts, was successfully crystallized and the structure solved at high resolution (1.65 A). The transporter was crystallized with substrate and Na+ bound to the transporter. The x-ray diffraction pattern revealed a protein containing 12 transmembrane segments (TMs) in a unique fold and with a binding site for L-leucine buried inside the center of the protein (Figure 14.3). The diffraction pattern also revealed an unexpected structural repeat in the first 10 TMs that relates TM1-5 with TM6-10 around a pseudo-twofold axis of symmetry located in the plane of the membrane. The binding pockets for leucine and Na+ are formed by TM1, TM3, TM6, and TM8. TM3 and TM8 are long helices that are related by the twofold symmetry axis and are strongly tilted (~50°) (Figure 14.3). TM1 and TM6 are characterized by unwound breaks in the helical structure in the middle of the lipid bilayer. These breaks expose main carbonyl oxygen and nitrogen atoms for direct interaction with the substrate.

The LeuTAa structure was crystallized in a conformation in which access to the substrate-binding site is closed from both the intracellular and extracellular environments, i.e., the predicted external and internal "gates" appear closed in the structure, and, hence, the structure likely represents an intermediate state between the outward facing conformation (where the substrate-binding site is exposed to the extracellular environment) and the inward facing conformation (where it is exposed to the intracellular milieu).

FIGURE 14.2 The alternating access model. The mechanism by which the neurotransmitter transporters translocate the substrate from the extracellular environment to the cytosol can be explained by the alternating access model. Without a neurotransmitter (NT) or ions (Na+, Cl-), the transporter resides in an outward facing conformation where the neurotransmitter-binding site is only accessible to the external environment. Upon binding of the solutes, the transporter undergoes a conformational change, first closing the outer gate excluding access to the binding site and subsequently opening the inner gate, allowing access to the binding site from the cytosol. The low Na+ concentration in the cytosol allows the release of the Na+ ions, which also causes a release of its substrate. The release of solutes again closes the inner gate and opens the outer gate, making the neurotransmitter transporter ready for another translocation cycle to occur. NT, neurotransmitter.

To accommodate shifts between outward- and inward-facing conformations it was suggested that the transport process involves major movements of TM1 and TM6 relative to TM3 and TM8. On the extracellular side, these movements were proposed to be controlled by an external gate that involves residues in TM1, TM3, TM6, and TM10, with a charged pair between Arg30 (TM1) and Asp404 (TM10) being of particular importance (Figure 14.3). Opening and closing of the external gate are also likely to involve conformational rearrangements of the extracellular loops (ECLs). ECL4 is especially interesting because it forms a lid that extends into the center of the transporter and interacts with residues in, for example, TM1. In this way, ECL4 covers the substrate-binding site from the outside without being in direct contact with the substrate (Figure 14.3). Thus, it is conceivable that opening the transporter to the outside involves a major conformational rearrangement of ECL4. This is in agreement with previous studies involving the engineering of Zn2+-binding sites and application of the substituted-cysteine accessibility method in the corresponding part of the DAT, SERT, and GABA transporter (GAT)-1.

The intracellular gate of LeuTAa is predicted to comprise ~20 A of ordered protein structure, involving in particular the intracellular ends of TM1, TM6, and TM8. A key residue in the predicted gate is Tyr268 at the cytoplasmic end of TM6, which is conserved in all transporters of this class (Figure 14.3). The tyrosine is positioned below the substrate-binding site at the cytoplasmic surface of the protein and forms a cation-n interaction with an arginine in the N-terminus just below TM1 that forms a salt bridge with an aspartate at the cytoplasmic end of TM8 (Asp369). A likely possibility would be that opening of the gate to the inside will require disruption of this set of interactions.

ecl4

2Na+

r^rPTM6.

FIGURE 14.3 High-resolution structure of Na+/Cl--coupled neurotransmitter transporter homologue (SLC6 family) from A. aeolicus (LeuTAa). (a) Schematic representation of transmembrane topology. The binding pocket for leucine and Na+ is formed by TM1, TM3, TM6, and TM8 (TMs highlighted in distinct colors; Na+ denoted as blue dots). (b) View of LeuTAa parallel to the membrane, highlighting important structural features. The central parts of TM1 (green) and TM6 (light blue) are characterized by unwound breaks in the helical structure. These breaks expose main carbonyl oxygen and nitrogen atoms for direct interaction with the substrate (red). The two Na+ ions (dark blue) also interact with the unwound part of TM1 and TM6 and have a key role in stabilizing this structure and the leucine-binding site. Access to the substrate-binding site from the external medium is predicted to be controlled by an external gate that in particular involves a charged pair between Arg30 (TM1) and Asp404 (TM10) as well as conformational rearrangements of the ECL 4 (yellow), which could form a lid that excludes the substrate-binding site from the extracellular space. A key residue in the predicted intracellular gate is Tyr268 at the bottom of TM6. This residue is conserved in all transporters of this class. Tyr268 interacts with Arg5 in the bottom of TM1, which again interacts with Asp369 at the bottom of TM8, with both residues also highly conserved throughout the SLC6 family. (c) View parallel to the membrane. (d) View from the extracellular side. The TMs forming the substrate-binding site are highlighted in distinct colors. (From Gether, U. et al., Trends Pharmacol. Sci, 27, 375, 2006. With permission.)

In agreement with this, recent experimental observations in the DAT in conjunction with computational simulations strongly support such a role of the interaction network and that the mechanism is highly conserved among all SLC6 transporters (Figure 14.3).

A major reorganization on the intracellular side during translocation is further supported by a marked increase in accessibility of cysteines introduced in the cytoplasmic half of TM5 of SERT to a membrane permeant cysteine-reactive reagent in the presence of serotonin. The observations indicated that when the transporter becomes inward-facing, the cytoplasmic half of TM5, which is occluded in the LeuT structure, lines an aqueous pathway leading from the binding pocket to the cytoplasm.

Taken together, based on the LeuTAa structure and available functional data, it seems reasonable to conclude that SLC6 transporters follow an alternating access model. However, the mechanism of transport by LeuTAa is probably distinct from those suggested for other ion-coupled transporters; hence, the mechanistic predictions clearly differ from those involving movements of two symmetrical hairpins reaching from the extracellular and intracellular environments, respectively, that were offered for sodium-coupled glutamate transporters based on a crystal structure of a bacterial member of this transporter family. Similarly, the suggested mechanism differs from the "rockerswitch" type mechanism proposed for Lac permease and the glycerol-3-phosphate transporter, two other recently crystallized transport proteins that mediate proton-coupled secondary active transport.

14.2.2 The Binding Sites for Na+ and Cl-: Importance in Substrate Binding and Translocation

The binding and cotransport of sodium ions is a feature that probably serves several purposes: first, the sodium ion(s) serve as a driving force for the translocation of the substrate against its electrochemical gradient; second, the ions coordinate the binding of the substrate to the transporter; and third, the ions might function as conformational guides, ensuring that the transporter undergoes the proper conformational changes during the translocation cycle.

SLC6 transporters bind and translocate one to three sodium ions during the translocation of one substrate molecule. In addition, several of the transporters bind and cotransport one chloride ion during one cycle, although this is not a ubiquitous feature of all transporters in the family. The SERT appears to be special because it also mediates the countertransport of one potassium ion during one transport cycle. The high-resolution structure of LeuTAa provided for the first time insight into the possible localization of the sodium-binding sites in SLC6 transporters by showing two distinct sodium-binding sites adjacent to the substrate-binding site. The two sodium ions appeared to have a key role in stabilizing the LeuTAa core, the unwound structures of TM1 and TM6, and the bound leucine molecule. One sodium ion (designated Na1) was found to possess an octahedral coordination, with one coordinate to the carboxyl group of leucine, thereby providing a possible structural link for the coupling of Na+ and solute fluxes. The other sodium ion (Na2) is positioned between the TM1 unwound region and TM8, about 7.0 A from Na1 and not directly involved in coordinating the bound leucine (Figure 14.3). Notably, with the conservative substitution of a serine for a threonine, all residues coordinating both Na1 and Na2 are conserved from LeuTAa to the mammalian transporters.

The LeuTAa does not possess any apparent Cl--binding site, and accordingly the transport of leucine is not dependent on the presence of Cl-. However, the use of homology modeling and energy minimization of the GAT-1 based on the LeuTAa structure, a potential chloride-binding site in this transporter, was elegantly identified. A cavity in the GAT-1 was found where the chloride ion may interact with the hydrogen atoms of the amide group of Gln291 and of the hydroxyl groups of Ser331, Ser295, and Tyr86 (GAT-1 numbering). The model was experimentally verified in part by the introduction of a negatively charged amino acid in position 331 (S331D/E), rendering both the net flux and the exchange of GABA largely chloride independent. Equivalent mutations introduced in the mouse GABA transporter-4 and the DAT also result in a chloride-independent transport, whereas the reciprocal mutations in LeuTAa and in Tyt1 convey these transporters from displaying chloride-independent substrate binding to chloride-dependent binding. Furthermore, the transport rate of GABA increased by lowering the intracellular pH, and thereby likely increasing the protonation during the return step of the glutamate inserted in position 331 of the GAT-1. This result suggests that in the wild-type transporter, the chloride ion is a substrate for the GAT and is released to the cytosol in contrast to simply binding to the protein throughout the entire translocation cycle. The requirement of the negative charge during the translocation of GABA, but not during the return step, suggests that the role of chloride is mainly to compensate for the multiple positive charges that enable accumulation of the substrate against huge concentration gradients.

14.2.3 Substrate Specificity and Binding Sites in SLC6 Neurotransmitter Transporters

The biogenic amine transporters include DAT, the norepinephrine transporter (NET), and the SERT. Among these, DAT and NET display marked overlapping selectivity for dopamine and norepineph-rine; hence, DAT transports dopamine and norepinephrine with similar efficacy, and the apparent affinity for dopamine is only a few folds higher than that for norepinephrine (Figure 14.4). Moreover, NET transports dopamine with 50% of the efficacy seen for norepinephrine, and the apparent affinity for norepinephrine is even a few folds higher than that for dopamine. Accordingly, their classification as DAT and NET seems primarily determined by their localization to dopaminergic and noradrenergic neurons, respectively, rather than by their distinct substrate specificity. In contrast, the SERT displays high specificity toward 5-HT although it has been shown that the SERT can transport dopamine if present in very high concentrations. The GABA transporters (GAT-1 to GAT-3) and the glycine transporters (GlyT-1 and -2) all display high specificity for their respective endogenous substrates; however, GAT-2 and GAT-3 can also transport beta-alanine, the only beta-amino acid that occurs naturally, and GlyT1 can transport the naturally occurring N-methyl-derivative of glycine, sarcosine.

The availability of the LeuTAa structure has provided the first reliable hypothesis for the location of the primary substrate-binding site in this class of transporters. The most striking feature of

Substrates

MPP+

Dopamine Norepinephrine Serotonin Amphetamine

MPP+

MDMA (Ecstacy)

Antagonists h=ulvochj

Cocaine

CH CH CH CH

Tropane class

MDMA (Ecstacy)

CH CH CH CH

JHW 007

GBR-12935 (Royal Gist-Brocades)

JHW 007

GBR-12935 (Royal Gist-Brocades)

Nr Mazindol

1,4-Dialkylpiperazine class Mazindol class

1,4-Dialkylpiperazine class Mazindol class

Citalopram Fluoxetine Paroxetine Methylphenidate

SSRIs Methylphenidate class

Nortriptyline X1: C, X2: =CH(CH2)2NHCH3

Tricyclic antidepressives

Citalopram Fluoxetine Paroxetine Methylphenidate

SSRIs Methylphenidate class

Duloxetine Venlafaxine

SNRIs

FIGURE 14.4 Chemical structures of most common substrates and antagonists for the DAT, NET, and SERT. SSRIs, selective serotonin reuptake inhibitors; SNRIs, serotonin-norepinephrine reuptake inhibitors.

Duloxetine Venlafaxine

SNRIs

FIGURE 14.4 Chemical structures of most common substrates and antagonists for the DAT, NET, and SERT. SSRIs, selective serotonin reuptake inhibitors; SNRIs, serotonin-norepinephrine reuptake inhibitors.

the leucine-binding site is that it exposes main-chain atoms from the unwound regions of TM1 and TM6 make most of the contacts with the a-amino and a-carboxy groups of the bound leucine. The unwound regions of TM1 and TM6 allow direct hydrogen-bonding partners as well as orientating the a-amino and a-carboxy groups so they can bind close to the ends of the helical segments and establish a-helix dipole interactions. In addition and as mentioned earlier, one of the Na+ ions makes direct contact with the carboxyl group of the leucine. Notably, this interaction cannot occur in transporters of the biogenic amines, dopamine, 5-HT, and norepinephrine, which do not possess a carboxyl group; however, the glycine in position 24 of LeuTAa (which is conserved among the amino acid transporters) is replaced with aspartate in these transporters (Asp79 in the DAT). According to the LeuTAa structure, this aspartate is predicted to be in the immediate proximity of Na+ and, thus, can probably substitute for the missing carboxyl group of the substrate. Studies on DAT and SERT also support the hypothesis that the aspartate in TM1 plays an important role in the coordination of the protonated primary amine in dopamine, and 5-HT. In GAT-1 the residue in this position is a glycine as it is in the LeuTAa. Hence, GABA is probably coordinated in the same way as leucine in the LeuTAa. Indeed, the residue seems to be important for the binding of GABA to the GAT-1.

Right above the leucine molecule in the LeuTAa structure, a tyrosine in TM3 (Tyr108) forms via its hydroxyl a hydrogen bond with the main-chain amide nitrogen of Leu25 in TM1. This interaction could function as a latch to stabilize the irregular structure near the unwound region in TM1 and may even be the first determinant of the closure of the extracellular gate. This hypothesis is even more interesting in light of the fact that the tyrosine is strictly conserved among all SLC6 family members and has been implicated in the substrate binding and transport of GAT-1 and SERT. Of further interest, recent homology modeling studies in DAT and SERT suggest that in these transporters, the hydroxyl group of the tyrosine does not form a hydrogen bond with the main-chain amide nitrogen of Leu25 but with the central aspartate in TM1, believed also to interact directly with the monoamine substrates (Asp79 in DAT and Asp98 in SERT and corresponding to position 24 in LeuTAa). In support of this hypothesis and thereby of the role of the hydrogen bond in stabilizing the substrate-binding site, mutation of the tyrosine in DAT (Tyr156) to phenylalanine decreases apparent dopamine around 10-fold and decreases the maximum uptake capacity by ~50%.

It is also important to note that the identification of residues shaping the leucine-binding site in LeuTAa illuminates the determinants of substrate specificity in the eukaryotic homologues. In the SERT, for example, residues at equivalent positions to those surrounding the isopropyl moiety of leucine in LeuTAa are replaced with smaller amino acids to accommodate the larger serotonin molecule. Correspondingly, homology modeling of the glycine transporters GluT1 and GlyT2 suggests together with mutational analysis that the substrate specificity is determined by a few key residues and that the ability of GlyT1 but not GlyT2 to transport sarcosine in addition to glycine is determined by a single residue difference between GlyT1 and GlyT2.

14.3 DRUGS TARGETING BIOGENIC AMINE TRANSPORTERS:

SPECIFICITY, USE, AND MOLECULAR MECHANISMS OF ACTION

The biogenic amine transporters, DAT, NET, and SERT, are targets for a wide variety of drugs. Overall, these drugs can be classified as either pure inhibitors that block substrate binding and transport, or as substrates that in addition to competing with the endogenous substrate are also transported themselves.

14.3.1 Cocaine, Benztropine, and Other Tropane Class Inhibitors

The most thoroughly studied class of inhibitors of biogenic amine transporters is the "tropane" class, with cocaine as the most well-known member (Figure 14.4). Cocaine is a moderately potent antagonist inhibiting the function of all three transporters nonselectively. However, earlier correlative studies as well as studies on genetically modified mice suggest that presynaptic DAT is the primary target for cocaine's stimulatory action. DAT knockout mice are insensitive to the administration of cocaine and, moreover, knockin mice expressing a DAT mutant incapable of binding cocaine shows insensitivity to cocaine administration. It is, therefore, the current view that the rapid increase in extracellular dopamine concentration elicited by cocaine inhibition of DAT produces the psycho-motor stimulant and reinforcing effect that underlie cocaine abuse.

Some closely related cocaine analogues possess higher potency toward the biogenic amine transporters and, thus, have been more suitable than cocaine itself in experimental setups (e.g., radioligand binding assays) directed toward understanding the pharmacological properties of the transporters. Important examples include CFT (2P-carbomethoxy-3P-(4-fluorophenyl)tropane or WIN 35,428) and RTI-55 ((-)-2P-carbomethoxy-3P-(4-iodophenyl)tropane or P-CIT) (Figure 14.4). Both compounds display nanomolar affinity for the biogenic amine transporters; however, while CFT shows selectivity for DAT over NET and SERT, RTI-55 shows selectivity for SERT and DAT over NET.

For compounds of the tropane class, the tropane ring and the 2P-carbomethoxy group are crucial for their affinity. An exception for this rule is the benztropine class. This group of tropanes lack the 2P-carbomethoxy group but still bind DAT with high affinity. Instead of the 2P-carbomethoxy group, the benztropines contain a diphenylmethoxy moiety. Recently, there has been increasing focus on benztropine analogues. Several of these compounds posses similar or even higher affinity and greater selectivity for the DAT than cocaine. The compounds tested so far readily cross the blood-brain barrier and produce increases in extracellular levels of dopamine for even longer durations than cocaine. Nonetheless, several of these DAT inhibitors are less effective than cocaine as behavioral stimulants. Furthermore, one benztropine analogue, JHW 007, has been found to potently antagonize the behavioral effects of cocaine (Figure 14.4). Assuming a correlation between behavioral effects of cocaine in laboratory animals and abuse potential in humans, these findings suggest JHW 007 as a potential lead for development of cocaine abuse pharmacotherapeutics. The reason for this discrepancy in the stimulating effect between cocaine and the benztropines has been suggested at least in part to be related to different pharmacodynamic properties of the compounds. Interestingly, recent studies suggest that while cocaine and cocaine analogues bind and stabilize an outward facing conformation of the transporter, benztropine analogues bind and stabilize a more closed conformation of the transporter. It is possible that binding to the open and likely more prevalent outward facing conformation of DAT conformation results in a faster on-rate, which may facilitate faster inhibition of DAT function and thereby a more rapid rise in extracellular dopamine concentration. In contrast, binding to a more closed and predicted less prevalent conformation of the transporter may result in a slower on-rate of the compound and thereby a slower rise in dopamine levels and a less stimulatory effect.

The molecular mode of interaction of cocaine and analogues with DAT has long been the subject of speculation. In particular, it has been debated whether or not the cocaine-binding site in DAT overlaps with that of dopamine. If inhibition of dopamine uptake by cocaine is the result of an allosteric mechanism, it would be possible, at least in theory, to generate a cocaine antagonist for treatment of cocaine addiction that might block cocaine binding without affecting dopamine transport. An experimentally validated molecular model of the cocaine-binding site in the DAT has been reported (Figure 14.5). The DAT model was generated on the basis of the LeuTAa structure followed by molecular docking of dopamine, cocaine, and other inhibitors into the model. The docking procedure revealed a binding site for cocaine and cocaine analogues that was deeply buried between TM1, 3, 6, and 8, and overlapped with the binding site for dopamine. There were, however, also significant differences between the binding modes: cocaine and cocaine analogues displayed a unique interaction with Asn157 and, moreover, the binding mode of cocaine/cocaine analogues distorted the conformation of the binding (Tyr156) that (as described earlier) plays a key role in closing the dopamine-binding pocket to the extracellular environment through formation of a stabilizing hydrogen bond with Asp79 in TM1 (Figure 14.5). In the cocaine-binding model, Tyr156 is pushed away by the 2P-methylester substituent of cocaine resulting in disruption of the hydrogen bond and a conformation of the binding pocket that is more open to the outside. This is in agreement with

Phe320 Ser321 Leu322 GLy323

The dopamine transporter (DAT)

Phe320 Ser321 Leu322 GLy323

The dopamine transporter (DAT)

(a)

DAT-dopamine TM8

DAT-CFT

N157

N157

N157

DAT-CFT

N157

FIGURE 14.5 Models of DAT/ligand complexes. (a) Two-dimensional schematic representation of the human dopamine transporter (hDAT). The colored circles denote residues that interact with either dopamine or the cocaine analogue CFT in the molecular models. Red circles, side chain interaction; orange, only backbone interaction. (b) Docked dopamine and (c) CFT in DAT. TMs 1, 3, 6, and 8 are shown in various shades of blue; the other TMs and intra- and extracellular loops have been removed for clarity. The ligands are shown in green. Sodium and chloride ions are shown as purple and salmon spheres, respectively. The protonated amine of dop-amine forms a salt-bridge with the Asp79 side chain (motif 1 in b). A polar interaction is also predicted between the amine of CFT and Asp79 (motif 1 in c). Dopamine further engages in hydrogen bonds with exposed backbone carbonyls of the unwound regions of TM1 and TM6. The binding of dopamine results in an additional aromatic-aromatic interaction between the catechol ring and Tyr156 (motif 2). Tyr156 also forms a hydrogen bond with Asp79 (motif 3) and a hydrophobic interaction with Leu80 (motif 4). In the CFT model, Tyr156 interacts in an edge-to-face manner with the methylester subtituent on the tropane ring (motif 5) leading to reorientation of the Tyr156 side chain and disruption of the hydrogen bond between Tyr156 and Asp79 seen in the dopamine model. Finally, there is CFT-specific hydrogen-bond interaction between the nitrogen of Asn157 and the fluoride atom of CFT (motif 6). (Modified from Beuming, J. et al., Nat. Neurosci., 11, 780, 2008. With permission.)

other results suggesting that cocaine locks the DAT in an outward facing conformation. The binding models have been substantiated by extensive mutagenesis of the proposed interacting residues. Specifically, the buried nature of the cocaine/cocaine analogue-binding site has been validated by trapping the radiolabeled cocaine analogue [3H]CFT in the transporter either through cross-linking of engineered cysteines or with an engineered Zn2+-binding site situated extracellular to the predicted common binding pocket.

In support of the proposed binding site for cocaine and benztropines, photoaffinity labeling studies in DAT with, e.g., the cocaine analogue [125I]RTI-82 and the benztropine analogue [125I]MFZ 2-24 have identified the possible major binding domains to be located in TM1 and 6, respectively. Although this technique does not permit detailed insight into the precise nature of the binding site, the results suggest that the binding sites are positioned in the same location as predicted by the binding models described earlier. In the SERT the most convincing data regarding the binding site for cocaine and cocaine analogues show that a combined mutation of Tyr95 in TM1 and Ile172 in TM3 markedly decrease the affinity for cocaine and RTI-55, suggesting that these residues possess a direct interaction very similar in the two transporters.

14.3.2 Amphetamine and Other Nonendogenous Substrates

Several nonendogenous compounds are substrates of the biogenic amine transporters and are used either as medication, drugs of abuse, or biochemical tools. Amphetamine and derivatives thereof, e.g., metamphetamine, p-chloroamphetamine, and 3,4-methylenedioxymetamphetamine (MDMA or ecstacy) are a class of psychostimulants that are transported by DAT, NET, and SERT (Figure 14.4). Methamphetamine preferentially acts on the DAT and NET while p-chloroamphetamine and MDMA have higher specificity for the SERT. This is supported by analyses of mice deficient in either DAT or SERT, i.e., DAT knockout mice are hyperactive and do not respond to amphetamine, while SERT deficient mice display locomotor insensitivity to MDMA. Interestingly, amphetamines do not only increase the synaptic concentration of dopamine DAT by competing with dopamine for uptake via DAT but also by promoting reversal of transport resulting in efflux of dopamine via the transporter. This efflux dramatically increases the levels of extracellular dopamine and is believed to be of major importance for the psychostimulatory properties of amphetamines. Increasing evidence supports that this efflux is not just the result of "facilitated exchange," but also might involve a channel mode of the transporter. Furthermore, recent studies suggest that the efflux is dependent on binding to the DAT C-terminus of Ca2+/calmodulin-dependent protein kinase a(CaMKIIa) that in turn facilitates phosphorylation of one or more serines situated in the distal N-terminus of the transporter.

As for the inhibitors, the binding sites for amphetamine and MDMA have also been investigated by molecular docking models suggesting an overlap of binding sites with dopamine.

14.3.3 Antidepressants

The biogenic amine transporters are also targets for medicines used against depression and anxiety. The selective serotonin-reuptake inhibitors (SSRIs), such as citalopram, fluoxetin, paroxetin, and sertraline, are, as implicated by their name, potent and selective inhibitors of the SERT (Figure 14.4). Another class of antidepressants includes the so-called serotonin-norepinephrine reuptake inhibitors (SNRIs) or "dual action" antidepressants such as venlafaxine and duloxetine that are active at both SERT and NET (Figure 14.4). Finally, the classical though still often used tricyclic antidepres-sants (TCAs) are potent inhibitors of NET and/or SERT with imipramine and amitriptyline being approximately 10 times more potent on SERT, while desipramine is relatively a selective inhibitor of the NET (see also Chapter 18). Interestingly, the antiobesity drug sibutramine exerts its action via combined inhibition also of NET and SERT. Conceivably, this effect is achieved through a combination of an anorectic effect due increased extracellular serotonin levels and increased thermogenesis due to increased norepinephrine levels.

The binding sites for antidepressants at their main targets: the NET and SERT are poorly described. Remarkably, it was recently reported that TCAs, such as clomipramine, imipramine, and desipramine, have activity at the bacterial homologue LeuTAa; hence, it was observed that the compounds were capable of noncompetitively inhibiting substrate binding to the transporter. The effect was only seen with high concentrations of the compounds and, thus, their affinity for LeuTAa is substantially lower than that observed for the NET/SERT. Nonetheless, it has been possibly to crystallize LeuTAa in complex with these compounds. The structures showed that clomipramine, desip-ramine, and imipramine bind in an extracellular-facing vestibule about 11 A above the occluded substrate-binding site, apparently stabilizing the extracellular gate in a closed conformation. The TCAs are cradled by the carboxy-terminal half of transmembrane helix 1 (TM1), the aminoterminal regions of TM6 and TM10, the approximate midpoint of TM3, and the sharp turn of ECL4.

The structures of TCA-bound LeuTAa obviously raise the key question whether they describe a binding mode for inhibitors that can be generalized to their mammalian counterparts. Mutations suggested that desipramine might interact in a similar fashion with the SERT; however, previous mutagenesis has supported that SSRIs, such as citalopram, fluoxetine, and sertraline as well as the TCA clomipramine, bind deeper in the transporter structure in a site more close to the substrate-binding site, i.e., mutation of Tyr95 in TM1 and/or Ile172 in TM3 of SERT substantially decreased the affinity for these compounds. Most significantly, the combined mutation of Tyr95 (Y95F) and Ile172 (I172M) decreased transporter affinity ~10,000-fold for escitalopram. The recent evidence for a buried binding site for cocaine and related inhibitors in DAT (see earlier) also strongly argues against that the TCA-binding mode seen in LeuTAa can be generalized to other transporters.

14.3.4 Other Biogenic Amine Transporter Inhibitors

The examples of additional biogenic amine inhibitors include the GBR (from Royal Gist-Brocades) analogues that are highly selective for DAT and mazindol (Figure 14.4) that inhibits NET with one and two orders of magnitude higher potency than DAT and SERT, respectively. Finally, the amphetamine derivative methylphenidate (Figure 14.4) is a potent blocker of primarily DAT and NET, and often used for treatment of narcolepsy and attention deficit hyperactivity disorder (ADHD). Not much is known about the molecular basis for the interaction of these compounds with the transporters.

14.4 INHIBITORS OF GLYCINE AND GABA TRANSPORTERS:

SPECIFICITY, USE, AND MOLECULAR MECHANISM OF ACTION

Other SLC6 family transporters than the biogenic amine transporters are targets for drugs or for drug discovery. GAT-1 is, for example, the target for the antiepileptic drug tiagabine; however, the molecular basis for its interaction with GAT-1 is not known. Recently, the N-dithienyl-butenyl derivative of N-methyl-exo-THPO (4-methylamino-4,5,6,7-tetrahydrobenzo[d]isoxazol-3-ol) (EF-1502) has been shown to inhibit not only GAT-1 but also the betaine carrier (BGT-1) and to act as a very efficient anticonvulsant whose action is synergistic with that of tiagabine (see also Section 15.4). Thus, BGT-1 is likely to be an important antiepileptic drug target. The explanation for the observations might be related to a differential distribution of BGT-1 and GAT-1. While GAT-1 is localized to synaptic sites, BGT-1 is localized to astrocytes and possibly extrasynaptic loci in the neurons; hence the efficacy of EF-1502 owing to its interaction with BGT-1 could be explained by modulation of extracellular GABA concentrations at extrasynaptic sites (for further details about GABA receptors and transporters see Chapter 15).

The high-affinity glycine transporters (GlyT1 and GlyT2) might also represent interesting drug targets. Physiologically, GlyT1 appears to play a role in astroglial control of glycine availability at NMDA receptors whereas GlyT2 is likely to play a fundamental role in glycinergic inhibition as reflected in a lethal neuromotor deficiency in GlyT2 knockout mice. The putative role of GlyT1 in regulating glycine availability at NMDA receptors has warranted attempts to develop high-affinity inhibitors of GlyT1 as a novel class of antipsychotic drugs, i.e., blockade of the GlyT1 is envisioned to increase synaptic levels of glycine ensuring saturation of the glycine-B (GlyB) site at the NMDA receptor at which glycine acts as an obligatory coagonist. Importantly, a derivative of sarcosine [3-(4-fluorophenyl)-3-(4'-phenylpheroxy)]propylsarcosine (NFPS) has been shown to potentiate NMDA receptor-sensitive activity and to produce an antipsychotic-like behavioral profile in rats. Several GlyT1 and GlyT2 inhibitors have now been described; however, little is known about their mode of interaction with the transporters.

14.5 CONCLUSION

The SLC6 neurotransmitter transporters represent a prototypical class of ion-coupled membrane transporters capable of utilizing the transmembrane Na+ gradient to couple "downhill" transport of Na+ with "uphill" transport (against a concentration gradient) of their substrate from the extracellular to the intracellular environment. The transporters play key roles in regulating synaptic transmission in the brain by rapidly sequestering transmitters such as dopamine, norepinephrine, serotonin, GABA, and glycine away from the extracellular space. Moreover, they are targets for a wide variety of drugs including antidepressants, antiepileptics, and psychostimulants as well as they are subject to current drug discovery efforts. Only recently, high-resolution structural information became available for this class of transporters through crystallization of the bacterial homologue, LeuTAa. For the first time, this permitted insight into the tertiary structure of this family of transporters. The structure serves as an important framework for future studies aimed at deciphering the precise molecular details and dynamics of the transport process. The structure also serves as an important template for delineating the molecular determinants for drug binding to SLC6 neurotransmitter transporters. The first experimentally validated computational models of drug binding have now been published and provided the first insight into the exact molecular basis for drug action at these important proteins.

FURTHER READINGS

Beuming, T., Kniazeff, J., Bergman, M. L., Shi, L., Gracia, L., Raniszewska, K., Newman, A. H., Javitch, J. A., Weinstein, H., Gether, U., and Loland, C. J. 2008. The binding sites for cocaine and dopamine in the dopamine transporter are overlapping. Nat. Neurosci. 11: 780-789. Gether, U., Andersen, P. H., Larsson, O. M., and Schousboe, A. 2006. Neurotransmitter transporters: Molecular function of important drug targets. Trends Pharmacol. Sci. 27: 375-383. Kniazeff, J., Shi, L., Loland, C. J., Javitch, J. A., Weinstein, H., and Gether, U. 2008. An intracellular interaction network regulates conformational transitions in the dopamine transporter. J. Biol. Chem. 283: 17691-17701.

Reith, M. E. A. 2002. Neurotransmitter Transporters: Structure, Function, and Regulation, 2nd edn. Humana Press, Totawa, NJ.

Singh, S. K., Yamashita, A., and Gouaux, E. 2007. Antidepressant binding site in a bacterial homologue of neurotransmitter transporters. Nature 448: 952-956. Yamashita, A., Singh, S. K., Kawate, T., Jin, Y., and Gouaux, E. 2005. Crystal structure of a bacterial homologue of Na+/Cl--dependent neurotransmitter transporters. Nature 437: 215-223. Yernool, D., Boudker, O., Jin, Y., and Gouaux, E. 2004. Structure of a glutamate transporter homologue from

Pyrococcus horikoshii. Nature 431: 811-818. Zomot, E., Bendahan, A., Quick, M., Zhao, Y., Javitch, J. A., and Kanner, B. I. 2007. Mechanism of chloride interaction with neurotransmitter:sodium symporters. Nature 449: 726-730.

Was this article helpful?

0 0
Natural Weight Loss

Natural Weight Loss

I already know two things about you. You are an intelligent person who has a weighty problem. I know that you are intelligent because you are seeking help to solve your problem and that is always the second step to solving a problem. The first one is acknowledging that there is, in fact, a problem that needs to be solved.

Get My Free Ebook


Post a comment