Báo cáo khoa học: A cocaine insensitive chimeric insect serotonin transporter reveals domains critical for cocaine interaction ppt

To delineate domains and residues that could play a role in cocaine interaction, the human serotonin transporter was mutated to incorporate unique amino acid substitutions, detected in t

A cocaine insensitive chimeric insect serotonin transporter reveals domains critical for cocaine interaction

Sumandeep K Sandhu1,2, Linda S Ross2and Sarjeet S Gill1,2

1

Environmental Toxicology Graduate Program and2Department of Cell Biology and Neuroscience, University of California, Riverside, USA

Serotonin transporters are key target sites for clinical drugs

and psychostimulants, such as fluoxetine and cocaine

Molecular cloning of a serotonin transporter from the

cen-tral nervous system of the insect Manduca sexta enabled us

to define domains that affect antagonist action, particularly

cocaine This insect serotonin transporter transiently

expressed in CV-1 monkey kidney cells exhibits saturable,

high affinity Na+and Cl–dependent serotonin uptake, with

estimated Kmand Vmaxvalues of 436 ± 19 nMand 3.8 ±

0.6· 10)18molÆcellÆmin)1, respectively The Manduca high

affinity Na+/Cl– dependent transporter shares 53% and

74% amino acid identity with the human and fruit fly

serotonin transporters, respectively However, in contrast to

serotonin transporters from these two latter species, the

Manduca transporter is inhibited poorly by fluoxetine

(IC50¼ 1.23 lM) and cocaine (IC50¼ 12.89 lM) To

delineate domains and residues that could play a role in

cocaine interaction, the human serotonin transporter was

mutated to incorporate unique amino acid substitutions,

detected in the Manduca homologue We identified a domain

in extracellular loop 2 (amino acids 148–152), which, when

inserted into the human transporter, results in decreased

cocaine sensitivity of the latter (IC50¼ 1.54 lM) We also constructed a number of chimeras between the human and Manducaserotonin transporters (hSERT and MasSERT, respectively) The chimera, hSERT1–146/MasSERT106–

587, which involved N-terminal swaps including trans-membrane domains (TMDs) 1 and 2, was remarkably insensitive to cocaine (IC50¼ 180 lM) compared to the human (IC50¼ 0.431 lM) and Manduca serotonin trans-porters The chimera MasSERT1–67/hSERT109–630, which involved only the TMD1 swap, showed greater sen-sitivity to cocaine (IC50¼ 0.225 lM) than the human transporter Both chimeras showed twofold higher serotonin transport affinity compared to human and Manduca sero-tonin transporters Our results show TMD1 and TMD2 affect the apparent substrate transport and antagonist sen-sitivity by possibly providing unique conformations to the transporter The availability of these chimeras facilitates elucidation of specific amino acids involved in interactions with cocaine

Keywords: serotonin; cocaine; antidepressants; transporter; Manduca

Among all the neurotransmitters, serotonin (also known as

5-hydroxytryptamine) remains historically the most

inti-mately involved with neuropsychopharmacology There is

ample evidence that the serotonin system modulates a

multitude of brain functions including sleep, mood,

cogni-tion, sensory percepcogni-tion, motor activity, temperature

regu-lation, nociception, appetite, sexual behavior and hormonal

secretion Disturbances in regulation of this system are

associated with severe behavioral malfunctions such as

depression, obsessive–compulsive disorder, and possibly

panic disorder, eating disorders, obesity and alcoholism [1,2]

The major mechanism by which serotonin action in the synaptic cleft is terminated is by its removal back into presynaptic nerve terminal via an uptake mechanism involving specific membrane transporters Widely pre-scribed antidepressant drugs like ProzacTM (fluoxetine) and ZoloftTM, which selectively inhibit this uptake, cause a profound increase in the concentration of serotonin at postsynaptic receptors and are used currently to treat various psychiatric disorders

Serotonin transporters have been cloned and functionally characterized from a number of organisms including human, rat, mouse and fruit fly [3–7] These transporters belong to a high affinity Na+/Cl–-dependent plasma-membrane transporters super-family The monoamine family of transporters that includes serotonin, dopamine and norepinephrine transporters (SERTs, DATs and NETs, respectively) share a high amino acid homology and display very distinct pharmacologies

These transporters are targets for the development of novel drugs Consequently, the analyses of structural and functional features of these transporters have captivated the interest of many researchers [8–16] Cloning of species variants and their comparative pharmacological studies

Correspondence to S S Gill, Department of Cell Biology and

Neuroscience, University of California, Riverside,

CA 92521, USA.

Fax: +1 909 787 3087, Tel.: +1 909 787 4621,

E-mail: Sarjeet.gill@ucr.edu

Abbreviations: TMD, transmembrane domain; EL, extracellular

loop; IL, intracellular loop; GABA, c-amino butyric acid;

SERT, serotonin transporter; NET, norepinephrine transporter;

DAT, dopamine transporter

Note: The sequence reported in this paper has been deposited in the

GenBank database (accession no AF384164).

(Received 2 April 2002, revised 12 June 2002, accepted 20 June 2002)

have made possible preliminary insights into the activity of

these transporters [10,11,17,18], thereby providing further

insights into designing more selective and hence safer

pharmacotherapeutics

As with vertebrates, serotonin has a well-defined role as a

neurotransmitter and neuromodulator in invertebrates [19]

The benefits of studying insect transporters are numerous,

not only providing potential targets for new insecticide

design, but also providing major insights into structure and

function analyses of these proteins In spite of being

evolutionarily distant, vertebrate and invertebrate

trans-porters show significant similarity at the primary protein

sequence level, and yet are very distinct functionally [6,7,20]

For example, the GABA (c-amino butyric acid) transporter

from Manduca sexta [20] possesses 58% identity to

mam-malian GABA transporter GAT1 and yet displays very

different sensitivities to classic GABA uptake inhibitors

We report here the molecular cloning and functional

characterization of the first lepidopteran serotonin

trans-porter (MasSERT) from the CNS of the tobacco hornworm

Manduca sexta.In comparison to mammalian homologues,

MasSERT has low sensitivity to cocaine and fluoxetine

Employing site-directed mutagenesis on the human SERT

(hSERT), and chimeras between MasSERT and hSERT,

this study also provides additional information regarding

cocaine recognition, possibly by governing conformational

changes in the transporter We anticipate that cloning of

MasSERT and the availability of hSERT/MasSERT

chi-meras will contribute to ongoing efforts of many researchers

in understanding the mechanism of action of

psycho-stimulants and antidepressants at the molecular level

E X P E R I M E N T A L P R O C E D U R E S

Materials

Vaccinia virus VTF-7)3was purified [21] using a crude stock

obtained from Dr Bernard Moss (National Institute of

Health, Bethesda, MD, USA) hSERT was kindly provided

by H Lester (California Institute of Technology, Pasadena,

CA, USA)

cDNA library screening

The full length MasSERT clone was isolated from a

size-selected (> 2 kb) cDNA library from the CNS of 5th

instar, day 3 larvae of M sexta The cDNA library was

constructed in pSPORT1 (Life Technologies, Gaithersburg,

MD, USA) as described previously [20] The cDNA library

was screened by the Limited growth PCR method [22]

Screening was carried out by nested PCR using

pSPORT1-specific and MasSERT-pSPORT1-specific oligonucleotide primers,

which were based on a partial MasSERT PCR product

(654 bp) obtained from Manduca sexta CNS The partial

PCR product was isolated by using synthetic degenerate

primers derived from short stretches of highly conserved

amino acid residues from the first (NVWRFPY) and sixth

(WIDAATQ) transmembrane domains of human

norepi-nephrine transporter (hNET) and rat GABA transporters

(rGAT1) as described previously [20] A full length

MasSERT cDNA was obtained subsequently, and

sequenced in both directions by the dideoxy chain

termin-ation method using an automated sequencer (Applied

Biosystems Inc.) Analysis of the nucleotide and deduced amino acid sequence was performed using the Lasergene software and programs from the Genetics Computer Group Sequence homology searches were done using BLAST

Expression construct

To express MasSERT in CV-1 cells, a NcoI/FseI fragment containing the MasSERT ORF was cloned into an expres-sion vector pTM1 [23] Similarly, hSERT was cloned into the pTM1 vector

Mutagenesis Comparison of MasSERT and hSERT sequences identified

a number of unique amino acid differences Specific point mutations were introduced into hSERT cDNA to convert these positions to give amino acids corresponding to the homologous sequences in MasSERT Mutations performed were Y134F, YM134–135FL, 189LA, 188A/189LA, FT191–192IN, F474Y, F515V, F551V (underlined residues indicate amino acid insertions) Mutagenesis was performed using the QuikChangeTM site-directed mutagenesis kit as described by the manufacturer (Stratagene) For each mutation, two complementary primers which contained the desired mutation were used (Table 1) All mutant clones were sequenced completely to confirm the presence of the mutation and the absence of any errors introduced by the polymerase

Chimera construction Chimeras were made by substituting segments of MasSERT cDNA for homologous segments in hSERT cDNA Six chimeras illustrated in Fig 1 were constructed hSERT cDNA was analysed for the presence of unique restriction enzyme cutting sites to be used as endpoints for the exchange of segments with MasSERT cDNA Once iden-tified, these restriction sites were engineered into homolog-ous sites of MasSERT cDNA This was accomplished

by synthesizing PCR products of MasSERT cDNA using primers homologous to corresponding regions of MasSERT cDNA, with the addition of the respective restriction site at the 5¢ end of each primer The KpnI site was in the vector DNA and therefore common to both MasSERT and hSERT The primers used for creating the restriction enzyme sites are shown in Table 2 The chimeras were sequenced completely to confirm chimera construction and the absence of any PCR errors

MasSERT antibody MasSERT immunoreactive serum was prepared by immunizing rabbits with an antigenic 14 amino acid peptide from the C-terminus of MasSERT and a N-terminal cysteine that was coupled to keyhole lymphet hemocyanin protein (Imject maleimide activated conjugation kit, Pierce) The MasSERT peptide corresponding to amino acids 571–84 (CQRPEVTSIPPADST) was synthesized by Research Genetics The crude serum was purified using peptide-coupled columns (Sulfolink kit, Pierce) and specific IgGs were aliquoted and stored at)80 C until needed

Detection of SERT expression

For Western blot analysis, total cell membranes were

prepared from the CNS of 5th instar M sexta, and

MasSERT cDNA and mock transfected CV-1 cells Tissues

from the CNS were homogenized in 10 vol of ice-cold

50 mM Tris/HCl (pH 7.4), containing 0.32M sucrose,

0.5 mM phenylmethanesulfonyl fluoride and protease

inhibitor cocktail (Sigma), and centrifuged at 3000 g for

10 min at 4C The supernatant was recentrifuged at

100 000 g for 1 h at 4C The resulting pellet was

suspended in 50 mMTris/HCl (pH 7.4) containing protease

inhibitors MasSERT and mock transfected CV-1 cell

membranes were similarly prepared following

homogeniza-tion in buffer [50 mM Tris/HCl (pH 7.4), 150 mM NaCl,

1% NP40, 0.5% deoxycholate, 0.1% SDS, 1 mM phenyl-methylsulfonyl fluoride, 1 mM dithiothreitol and protease inhibitor cocktail] Samples obtained were incubated at

42C for 30 min in sodium dodecyl sulfate (SDS) sample buffer (2% SDS; 62 mMTris/HCl, pH 6.8; 10% glycerol; 0.77% dithiothreitol; 0.01% bromophenol blue) and separ-ated by SDS/PAGE using 3% stacking and 10% resolving gels The gels were transferred to Immobilon-P membranes (Millipore corporation) by standard procedures [24] The membranes were then treated with blocking buffer (1· NaCl/Pi; 13.7 mM NaCl, 0.27 mM KCl, 0.43 mM

Na2HPO4Æ7H2O, 0.14 mM KH2PO4), 1% bovine serum albumin, 0.05% Tween (20) for 1 h at room temperature on

a shaker The membrane was then incubated overnight at

4C with MasSERT antibody, diluted to 1 : 1500 in

Table 1 Primers used for generating mutations.

Name of Mutant Mutation Primers

5¢-CCAGTGCGAGCTCCATGAAAAAGAGCGGGATTCCCCC (reverse) YM134–135FL M to L 5¢-CCCGCTCTTTTTCCTGGAGCTCGCACTGGGAC (forward)

5¢-GTCCCAGTGCGAGCTCCAGGAAAAAGAGCGGG (reverse) 189LA Insert L+A 5¢-CGCTATACTACCTCATCTCCTTAGCTTCCTTCACGGACCAGCTGC (forward)

5¢-GCAGCTGGTCCGTGAAGGAAGCTAAGGAGATGAGGTAGTATAGCG (reverse) 188A/189LA Insert A 5¢-GCGCTATACTACCTCATCGCTTCCTTAGCTTCCTTCACG (forward)

5¢-CGTGAAGGAAGCTAAGGAAGCGATGAGGTAGTATAGCGC (reverse) YM134–135FL FT to IN 5¢-CTACCTCATCTCCTCCATCAACGACCAGCTGCCCTGGAC (forward)

5¢-GTCCAGGGCAGCTGGTCGTTGATGGAGGAGATGAGGTAG (reverse)

5¢-GTCAGGGTGACCAGGGATCCAAAGTAGCAGGTGATGACCA (reverse)

5¢-CCCTGCAGAACTGAGTGATGCCATAGACCCAAGACACAGCGAC (reverse)

5¢-GGCTCATCAGAAAACTGCAAATGATGACCAGGAGAAACAGAGGGC (reverse)

Fig 1 Hydrophobicity based models illustra-ting chimeras of MasSERT and hSERT Six chimeras were constructed as described in Experimental procedures All replacements represent exchanges with homologous regions

of the respective cDNAs Numbers refer to amino acids sequence of the respective transporters (A) MasSERT(1-67)/

hSERT(109-630); (B) MasSERT(1-291)/ hSERT(333-630), (C) hSERT(1-108)/ MasSERT(68-291)/hSERT(333-630), (D) hSERT(1-146)/hSERT(106-291)/ hSERT(333-630), (E) hSERT(1-146)/ MasSERT(106-587), (F) hSERT(1-333)/ MasSERT(293-587) Chimeras A and E were functional.

blocking buffer As a control, another membrane

contain-ing the same samples was incubated with MasSERT

antibody preadsorbed with a 30-fold excess of the

MasSERT peptide at 4C for 16 h The immunoreactivity

was detected using horseradish peroxidase-coupled donkey

anti-(rabbit IgG) secondary Ig, in combination with the

ECL detection system (Amersham)

In vitro translation

In vitrotranslation of MasSERT was performed using a

TNT rabbit reticulocyte lysate kit (Promega) according to

the manufacturer’s instructions using [35S]methionine The

translated products were separated by SDS/PAGE as

described above The gel was then stained with Coomassie

Brilliant Blue G-250, destained and treated with Entensify

(NEN Research Products) as per manufacturer’s

instruc-tions and exposed to HyperfilmTM MP autoradiography

film (Amersham Life Science) at)70 C for 2 days

Expression in CV-1 cells

CV-1 cells were maintained following standard procedures

[21] Transient expression of MasSERT and hSERT in

CV-1 cells was carried out using recombinant vaccinia virus

VTF-7)3expression system as described previously [25]

Transport assays

Transfected cells were washed with KRTH buffer (10 mM

Hepes, pH 7.4, 120 mMNaCl, 4.7 mMKCl, 5 mMTrisHCl,

5 mM KH2PO4, 2 mM CaCl2, 1.2 mM MgSO4, 5.6 mM

glucose, 100 lM L-ascorbate, 100 lM pargyline) [25] and

incubated with KRTH buffer for 10 min For transport

studies, cells were then incubated with either [3H]serotonin

(10.2 CiÆmmol)1, Amersham) alone or as a mixture of

unlabeled and [3H]serotonin at the concentrations indicated

After a 15 min incubation at room temperature, the cells

were washed twice with cold uptake buffer, then solubilized

in 1% SDS and the radioactivity of cell extracts was

measured by liquid scintillation counting Inhibition studies

were carried out similarly in the presence of varying

concentrations of inhibitors and a constant amount of

[3H]serotonin (0.05 lM) for 15 min at room temperature For studies with cocaine, the esterase inhibitor, phenyl-methylsulfonyl fluoride, was used at a concentration of

100 lMin the uptake assay mixture to prevent degradation

of cocaine All incubations were carried out in duplicate, and experiments were replicated a minimum of three times Because no difference was observed between nonspecific [3H]serotonin uptake levels in cells transfected with the pTM1 vector containing the M sexta GABA transporter and cells transfected with no DNA (mock transfected), we chose to use mock-transfected cells as a negative control in all our experiments Nonspecific uptake was defined in parallel wells in duplicates and was subtracted from the total uptake to yield the specific uptake All data represent specific uptake The reported Km and IC50 values were obtained by analysing the data inORIGIN(MicroCal Inc.) using the Levenberg–Marquardt algorithm and by fitting the curves using the simplex method for non-linear least squares

R E S U L T S

The MasSERT cDNA clone isolated was 3717 bp in length Based on the consensus start site sequence [26] and the longest ORF, the start methionine is predicted to be at position 151 bp The ORF is 1764 bp with a deduced amino acid sequence of 587 amino acids, and a 1803 bp 3¢ UTR The hydrophobicity profile of MasSERT indicates the presence of 12 putative transmembrane domains (TMD), characteristic of this superfamily of proteins There are two putative N-glycosylation sites, one in the large extracellular loop between TMD 3–4 (amino acids 182–85) and another between TMD 7–8 (amino acids 350–53) The unglycosy-lated transunglycosy-lated MasSERT is predicted to have a molecu-lar mass of 64.8 kDa However, the in vitro translated MasSERT migrates as a 45-kDa band in SDS/PAGE whereas native and heterologously expressed MasSERT migrated as 55 kDa and 90 kDa bands suggesting possible glycosylation of MasSERT and/or existence of dimers, respectively (Fig 2) An additional 45 kDa band was detec-ted in CV-1 cells transfecdetec-ted with MasSERT cDNA that may account for the unglycosylated form of MasSERT (Fig 2) Interestingly, no putative protein kinase C phosphoryla-tion sites were predicted in the MasSERT, whereas seven proline-directed protein kinase phosphorylation sites were predicted in the regions in cytoplasmic domains based on the proposed topology modeled for GAT1 [8,9] Phospho-rylation of hSERT by protein kinase C results in a reduction

in the number of transporters on the cell surface [27], implying MasSERT might be regulated differently The N-terminus also contains three PXXP motifs that could bind SH3 domains [28], which play an important role in signal transduction

Sequence comparisons with other known members of the family indicate that it indeed is a member of Na+/Cl– -dependent neurotransmitter transporter family Dendro-gram analyses based on sequence alignment with previously reported transporters show that it is most closely related to serotonin transporters (Fig 3) MasSERT displays 53% and 74% amino acid identity to hSERT [5] and dSERT [6,7], respectively

Transient expression of MasSERT in CV-1 cells showed significant (15–40·) increase in cellular [3H]serotonin levels

Table 2 Primers used for generating chimeras.

Restriction

site Primers

(forward)

(forward)

5¢-CGCCATATGTAGGGGAATCGCCACACGT

(reverse)

NsiI 5¢-ATAATGCATCACTCTCTGGAAACGGATC

(forward)

BglII 5¢-TTAGATCTTCTTCTCGCTCGGTCCCGG

(forward)

5¢-TTAGATCTGGGATGCCGCGTCAATCC

(reverse)

FseI 5¢-CAGTACGGCCGGCCTCACAGGTT (reverse)

as compared to the background levels of mock-transfected

cells MasSERT cDNA-transfected cells did not transport

radiolabeled GABA, glycine, proline, glutamate or leucine

above the background levels (data not shown) MasSERT

showed strong substrate specificity for serotonin over other

biogenic amine substrates including dopamine,

norepineph-rine, octopamine, histamine and tyramine (Fig 4A,

Table 3) At 10 lM concentration, serotonin uptake in

MasSERT-transfected CV-1 cells is linear until 20 minutes

and reached a plateau at 45 minutes (Fig 4B) The transport

was saturable, which indicated the expression of a

carrier-mediated uptake system (Fig 4C) The Michaelis Menton

constant (Km) for serotonin uptake was 436 ± 19.2 nM

(n¼ 5) with a Vmaxof 3.84 ± 0.61· 10)18molÆcellÆmin)1

Similar Kmvalues for serotonin transport were determined

for cloned and endogenous SERTs [4,5,29]

The high affinity transport of serotonin by MasSERT

was dependent on extracellular Na+and Cl–ions (Fig 4D)

Substitution of Na+with choline and substitution of Cl–

with acetate or gluconate in the transport buffer totally

eliminated serotonin transport Chloride ion requirement

for MasSERT is thus different from dSERT, where Cl–

facilitates serotonin transport but is not an absolute

requirement [7] In contrast to mammals, Manduca

hemo-lymph is characterized by a high K+)Na+ ratio [30]

Several species of phytophagous lepidopteran larvae have been reported to possess K+-coupled transport systems [31] Based on these observations, we also tested the effect of extracellular K+ concentration in driving the serotonin transport by MasSERT Addition of 120 mM K+ to the transport buffer as well as a complete depletion of K+from the transport media did not significantly affect serotonin transport into CV-1 cells (Fig 4D), suggesting MasSERT is not a K+-coupled transporter

The pharmacological sensitivity of MasSERT to a variety

of well characterized serotonin uptake blockers and sub-strates is shown in decreasing rank order of potencies in Table 3, along with the comparison of inhibition constants reported elsewhere for hSERT Among the antagonists tested in this study, mazindol is the most potent inhibitor of MasSERT with an IC50of 153 nMthat is similar to the IC50 determined for hSERT but 40· less potent than that for dSERT and hDAT [6,7,32] Mazindol is a potent inhibitor

of norepinephrine and dopamine transporters with IC50 values of 1 nMand 11 nMfor hNET and hDAT, respect-ively [32,33]

Nomifensine, a selective norepinephrine uptake blocker, was an extremely weak inhibitor of MasSERT, with an IC50 value of 7.9 lM However, clomipramine and desipramine, two other tricyclic antidepressants, were better inhibitors with IC50 values of 370 and 638 nM, respectively Trypt-amine, which is a substrate for the endogenous platelet SERT, inhibited half maximal MasSERT-mediated sero-tonin uptake at 5.3 lM Other potential substrates, octop-amine, dopoctop-amine, norepinephrine, tyroctop-amine, tryptophan and histamine did not inhibit MasSERT at concentrations

up to 200 lM Fluoxetine, an effective selective serotonin uptake inhi-bitor in clinical use, and imipramine, the tertiary amine tricyclic antidepressant drug, were very weak antagonists of MasSERT with IC50 values of 1.23 lM and 1.76 lM, respectively Similarly, cocaine, the most actively studied non-selective inhibitor of biogenic amine transporters and a psychostimulant, was unable to inhibit MasSERT-mediated serotonin transport at concentrations sufficient to inhibit half maximal transport of mammalian serotonin, dopamine and norepinephrine transporters To confirm our findings,

we performed inhibition assays with MasSERT and hSERT under identical experimental conditions Our results showed similar inhibition profile of hSERT-mediated serotonin uptake by cocaine and fluoxetine as reported previously [5,7], with IC50values of 431 nM and 4.2 nMrespectively, but a very weak inhibition of MasSERT-mediated serotonin uptake (Table 3, Fig 5)

PILEUPanalysis of MasSERT with other members of the monoamine transporter family (Fig 3) indicated that, despite having a high sequence identity to hSERT and dSERT, there are amino acid residues that are unique to MasSERT Because MasSERT is 30· less sensitive to cocaine than hSERT, the role of these unique amino acid sequences in cocaine sensitivity was analysed Our initial focus was on tyrosine and phenylalanine residues because aromatic rings of these amino acids form polar p–p stacking

or cation–p interactions with aromatic ligands [34] Studies with nicotinic acetylcholine [35] and tachykinin receptors [36] showed that substitution of these amino acids with one another at functionally important sites is not always tolerated Moreover, cocaine analogues lacking phenyl

Fig 2 In vitro translation and Western blot analysis using affinity

purified MasSERT antibody In vitro translated MasSERT (lane 1)

runs as a 45-kDa protein on SDS/PAGE For western analysis, equal

amounts of cell membranes (25 lg per lane) from Manduca sexta

CNS, mock-transfected CV-1 cells and MasSERT-transfected CV-1

cells (lanes 2–4, respectively) were subjected to SDS/PAGE and

im-munoblotted with affinity purified anti-MasSERT Ig as described in

Experimental procedures The anti-MasSERT Ig recognized 55 and

90 kDa bands in the CNS and MasSERT-transfected cells An

addi-tional immunoreactive 45 kDa band is detected in transfected cells.

The 55 and 90 kDa proteins are likely to represent glycosylated

MasSERT and/or dimers, respectively, with the 45 kDa being the

unglycosylated form of MasSERT This immunoreactivity can be

com-peted when the MasSERT antibody is preadsorbed with MasSERT

peptide (lanes 5–7 were loaded similar to lanes 2–4) showing that the

interaction of MasSERT antibody is specific to MasSERT.

Fig 3 Amino acid sequence alignment of MasSERT with known monoamine transporters Deduced amino acid sequences of hSERT [5], dSERT [6,7], hDAT [32], hNET [33], dDAT [40] were aligned using the GCG program Identical amino acids are shown on a black background, while conserved residues are shaded The putative TM membrane spanning domains are overlined.

Fig 4 Characterization of MasSERT-mediated [ 3 H]serotonin uptake in CV-1 cells (A) Substrate specificity CV-1 cells transfected with MasSERT cDNA were incubated with 0.1 l M3H-labelled substrates for 15 min as indicated The data represents percentage of substrate uptake above control levels (B) Time course of serotonin transport CV-1 cells transfected with MasSERT cDNA were incubated with 10 lm serotonin for the indicated time (C) Kinetics of serotonin uptake MasSERT-transfected cells were incubated with [3H]serotonin and increasing concentrations of unlabeled serotonin for 15 min as described in Experimental procedures The Eadie–Hofstee analysis is depicted in the inset of the figure (K m ¼ 439 n M ,

V max ¼ 3.3 · 10)18mol per cell per min) The data represents specific serotonin transport, expressed as fmol per cell per hour, and is from a single experiment that was repeated 4 more times with similar results (D) Ion dependence MasSERT cDNA-transfected CV-1 cells were incubated for

10 min with 50 n M [3H]serotonin Non-specific uptake was determined in CV-1 cells transfected with no DNA and subtracted from each determination The data are given as percentage of specific uptake above control levels using values from duplicate wells The original assay buffer was changed according to the different ions tested For the assay of cations, NaCl was replaced by equimolar concentration of choline chloride For the assay of anions, chloride was replaced by equimolar salts of sodium and potassium gluconate and sodium and potassium acetate For potassium dependence, the assay was done either in the buffer containing no KCl and KH 2 PO 4 , or buffer containing 120 m M KCl This buffer also contained

120 m NaCl.

rings have extremely low affinity for the dopamine

trans-porter [37,38]

Therefore a number of mutations were performed by

focusing on unique aromatic substitutions in hSERT

by substituting with the amino acid residues found in

MasSERT at the corresponding positions Thus the

mutants Y134F, YM134–135FL, FT191–192IN, F474Y, F515V, F551V were made In addition two more mutants in the EL2, 189LA and 188A/189LA, were made that had three amino acid insertions, which extend that particular region of MasSERT EL2 when compared with the rest of the neurotransmitter transporter superfamily

All of these mutants were functional All mutations involving aromatic residues showed no statistically signifi-cant difference in cocaine sensitivity compared to hSERT (Fig 6A, Table 4) However, the mutant Y134F showed a lower level of sensitivity to cocaine than hSERT The two EL2 mutants, 189LA and 188 A/189LA, that had amino acid insertions, were inhibited at relatively higher concen-trations of cocaine with IC50values of 1163 ± 31 nMand

1542 ± 42 nM, respectively (Fig 6B, Table 4) These val-ues are statistically significant at P < 0.005

Because these initial mutations did not result in major changes in cocaine sensitivity of hSERT we then construc-ted chimeras of MasSERT and hSERT Six chimeras were constructed making use of available restriction sites

in hSERT (Fig 1) Only two of the six chimeras, Mas-SERT(1-67)/hSERT(109-630) and hSERT(1-146)/Mas-SERT(106-587) were functional The MasSERT(1-67)/ hSERT(109-630) chimera displayed a twofold increase in cocaine potency (IC50¼ 225 ± 11 lM) as well as higher substrate transport specificity (231 ± 11 nM, P < 0.05 versus hSERT) for serotonin (Fig 6C, Table 4) compared

to wild type hSERT The hSERT(1-146)/MasSERT(106-587) chimera also had higher specificity for serotonin (197 ± 19.nM, P < 0.05 versus hSERT) but a dramatic decline in cocaine sensitivity (Fig 6C, Table 3), compared not only to hSERT but also to MasSERT, exhibiting an

IC50 value of 180 ± 8.3 lM (n¼ 3) These differences in cocaine sensitivity are statistically significant at P < 0.005 compared to both hSERT and MasSERT The substrate saturation experiments for MasSERT, hSERT and chi-meras were done at same time under identical kinetic

Table 3 Pharmacological specificity of [ 3 H]serotonin uptake in CV-1 cells transfected with MasSERT cDNA IC 50 (n M ) values for inhibition of [3H]serotonin uptake for various antagonists and substrates are listed in accordance with their rank order of potency CV-1 cells transfected with MasSERT cDNA were incubated for 15 minutes with [ 3 H]serotonin and various concentrations of the indicated compounds Data represents the mean ± SEM of 3–5 independent experiments, each conducted in duplicate For all experiments, [ 3 H]serotonin concentration was kept constant at 0.05 l M IC 50 values for hSERT are also included from our study and/or previously published reports ND, not determined.

73 ± 5.6a

a Demchyshyn et al., 1994, b Ramamoorthy et al., 1993, c Barker et al., 1994.

Fig 5 Fluoxetine inhibition of the [ 3 H]serotonin uptake into CV-1 cells

transfected with MasSERT and hSERT Transfected cells were

incu-bated with fluoxetine at the indicated concentrations for 15 min The

[ 3 H]serotonin concentration was kept constant at 50 n M Non-specific

uptake was subtracted from the total uptake to yield specific

[ 3 H]serotonin uptake The data is presented as the percentage mean

values of [ 3 H]serotonin uptake in the absence or presence of the

an-tagonists The IC 50 values obtained from the inhibition curves are:

fluoxetine, hSERT-3.7 n M , MasSERT-1.08 l M The data represented

is from a single experiment that was repeated at least five times with

similar results.

conditions The Kmvalues for hSERT and MasSERT are

460.5 ± 17.5 nMand 436 ± 19.2 nMrespectively

D I S C U S S I O N

MasSERT described here is the only known Na+/Cl–

-dependent serotonin transporter that displays a significant

relative insensitivity to psychostimulants like cocaine and

the antidepressant fluoxetine However, in spite of the low

sensitivity to cocaine, MasSERT has serotonin transport

affinity similar to that observed with other SERTs All

mammalian and nonmammalian monoamine transporters

identified so far are cocaine sensitive and have comparable

inhibition constants [4–7,32,33,39] except for DATs from

Drosophila melanogaster [40] and Caenorhabditis elegans

[41], which have reported IC50values of 2.6 lMand 5 lM,

respectively The IC50 values for human, mouse, rat and

fruit fly SERTs reported by various groups fall in the

range of 300–600 nMwith minor differences attributable to

differences in the experimental conditions In contrast

MasSERT shows no sensitivity to cocaine in this concen-tration range, but displayed 30 times less sensitivity than human SERT and DAT [4,5,32,42] MasSERT was 300–

400 times less sensitive to fluoxetine than human SERT, for which fluoxetine is a potent inhibitor (IC50¼ 3–5 nM); it is also less sensitive than Drosophila SERT (IC50 73 nM) This difference in pharmacology is not entirely due to differences between insects and mammals, as the Drosophila SERT shows high sensitivity to both cocaine and fluoxetine However, MasSERT is nearly equally sensitive as dSERT

to the tertiary amine tricyclic antidepressants, imipramine and desipramine These differences in pharmacology sug-gest MasSERT is likely to have unique structural domains, compared to other SERTs, making it insensitive to cocaine and fluoxetine

Cocaine abuse in the United States continues to remain a major socioeconomic and medical issue of modern society, with no effective treatment available for cocaine dependence [14,43] DATs, SERTs and NETs are major targets for the reinforcing actions of cocaine [12,44–47] It is known that

Fig 6 Cocaine inhibition of the [3H]serotonin

uptake into CV-1 cells transfected with hSERT,

mutant hSERT and chimeras Transfected cells

were incubated with cocaine at the indicated

concentrations for 15 min The [ 3 H]serotonin

concentration was kept constant at 50 n M

Unless otherwise specified, the data is

presen-ted as the percentage mean values of

[3H]serotonin uptake in the absence or

pres-ence of the antagonists Non-specific uptake

was subtracted from the total uptake to yield

specific [3H]serotonin uptake The data

rep-resented are from a single experiment that was

repeated at least three times with similar

results The mean IC 50 values obtained from

these inhibition curves are given in Table 4.

(A) and (B) hSERT with specific amino

acid mutants (C) hSERT and MasSERT

chimeras.

Table 4 Cocaine inhibition of [ 3 H]serotonin in mutant hSERT and MasSERT/hSERT chimeras CV-1 cells transfected with wild type or mutant cDNAs were incubated for 15 minutes with [3H]serotonin and increasing concentrations of cocaine Data represents the mean ± SEM of 3–4 independent experiments, each conducted in duplicate For all experiments [ 3 H]serotonin concentration was kept constant at 0.05 lM The Student’s t-test was performed for statistical analysis of IC 50 values compared to hSERT P** < 0.005, P* < 0.02 The mutants F474Y, F515V, F551V also did not show substantial change in cocaine sensitivity The K m and V max values for MasSERT, hSERT and chimeras are derived from Eadie–Hofstee analysis of kinetics of [ 3 H]serotonin transport.

Transporter Protein Cocaine IC 50 (n M ) K m (n M ), V max (mol per cell per min)

dopamine and serotonin transporters have distinct domains

for substrate recognition and antagonist binding [11,18,

44,48,49] Experiments involving protection of hDAT

regions from alkylation with N-ethylmaleimide substrates

show differential binding of dopamine and cocaine [49]

Cocaine and benztropine bring differential conformational

changes in hDAT that makes amino acid C90 available to

methanethiosulfonate reagents only in the presence of

cocaine [50] In spite of major advances made in this field

using chimera construction and mutagenesis [8–10,13–

15,17,43,51,52], a cocaine-binding site on the serotonin or

dopamine transporters has not been resolved These studies

have provided evidence for a possible role of TMD 1–2, 4–5,

8, 11 and 12 in cocaine recognition However, due to a lack

of cocaine selectivity among the monoamine transporters,

the conclusions from these studies are based on small (2–8·)

differences observed in cocaine potency

Our human SERT mutants, Y134F, YM134–135FL,

FT191–192IN, F474Y, F515V, and F551V did not show

any change in cocaine sensitivity compared to the wild type

Possibly these amino acid changes, which are unique to

MasSERT, do not interact with cocaine but could

contri-bute towards recognition and binding of other SERT

antagonists Phenylalanine and threonine in EL2

[FT(191-192)] are absolutely conserved in all monoamine Na+/Cl–

dependent transporters, except in MasSERT where they are

substituted by isoleucine and asparagine, however, these

residues are not important for cocaine sensitivity Similarly

YM134-135 in hSERT (FL in MasSERT) is absolutely

conserved in amine transporter subfamily Mutations at

these amino acids also did not affect hSERT sensitivity to

cocaine

Although the extracellular loops (EL) between

trans-membrane domains do not appear to be responsible for

substrate specificity and antagonist selectivity, these loops

may provide the desired conformation required for proper

transporter function [53,54] It is interesting to note that

EL2 in MasSERT carries two additional amino acid

residues A148 and S149, which extend this region of EL2

as compared to rest of the superfamily The hSERT

mutants, 189LA and 188A/189LA, mutagenized to

intro-duce the corresponding region, were less sensitive to cocaine

(Fig 6B, Table 4) Although the mutant 188A/189LA did

not show a dramatic shift in cocaine potency towards

MasSERT, it might be sufficient to bring subtle

conforma-tional change in the transporter or even make one of the

many possible recognition sites for cocaine

As functional data from specific mutations in hSERT did

not yield sufficient information to explore the binding sites

for cocaine, we focused our attention on the classical

approach of constructing chimeras between hSERT and

MasSERT, using available restriction sites in hSERT In

chimera hSERT(1–146)/MasSERT(106–587) insertion of

the N-terminus of hSERT in MasSERT, by replacing its

first 105 amino acids, makes it 418· and 14· more resistant

to cocaine than hSERT and MasSERT, respectively It was

expected that this chimera would either behave similarly to

MasSERT in the presence of cocaine or its cocaine

sensitivity curve would shift towards hSERT Because this

chimera was much more resistant to cocaine and yet had an

improved transport affinity to serotonin, it makes an

excellent tool to identify domains and amino acid residues

which could be potentially involved in cocaine interaction

Interestingly, chimera MasSERT(1–67)/hSERT(109–630), which only contains the N-terminal 67 amino acids of MasSERT, was more sensitive to cocaine than hSERT, and

it displays similar higher transport affinity to serotonin as observed for chimera hSERT(1–146)/MasSERT(106–587) These results suggest that the N-terminus, including TMD1–2, plays a substantial role in providing a unique conformation to the transporter thereby governing the substrate transport affinity, cocaine sensitivity and possibly sensitivities to other antagonists Based on chimera design, it appears that TMD 1 of MasSERT and TMD1–2 of hSERT contain unique molecular determinants that interact differ-entially with the rest of the transmembrane domains of hSERT and MasSERT, respectively Previous studies with cross-species chimeras have provided evidence that TMD1–

2 might play a critical role in antagonist recognition [10,52] However, these chimeras displayed marginal or no differ-ences for cocaine potencies For example, chimeras con-structed between hSERT and dSERT at similar positions, dSERT(1–136)/hSERT(137–625) and hSERT(1–118)/ dSERT(119–627), [10] exhibit comparable potencies for cocaine to those found in the parental transporters Similarly, DAT and NET chimeras that intersect within

or near TMD1 have been shown to have only slightly lower potency for cocaine than wild type DAT and NET [52] The availability of transporters and chimeric transporters having

a wide range of sensitivities to cocaine (225 nMto 180 lM) facilitates a systematic probe of structural determinants Efforts are underway to further investigate the pharmaco-logical properties of these two chimeras in order to precisely define the domains/amino acid residues important for bringing conformational changes to the transporter and antagonist binding

Taken together, it is evident that MasSERT is compar-atively less sensitive to cocaine and other pharmacological agents than most members of the monoamine transporter subfamily Future studies exploiting the pharmacologi-cal differences found in MasSERT and chimeras hSERT (1–146)/MasSERT(106–587) and MasSERT(1–67)/hSERT (109–630), coupled with rational site-directed mutagenesis

of MasSERT and hSERT may contribute to our present understanding of domains that dictate drug selectivity The availability of MasSERT and chimera hSERT(1–146)/ MasSERT(106–587) could contribute towards understand-ing cocaine action

A C K N O W L E D G E M E N T S

We thank Henry Lester, Caltech, USA, for providing the human serotonin transporter cDNA and Valery Filippov for the helpful discussion and advice during the course of this work and the preparation of the manuscript This research was supported by grants from the NIH (AI 34524 and AI 48049 to S S G.).

R E F E R E N C E S

1 Nemeroff, C.B (1998) Psychopharmacology of affective disorders

in the 21st century Biol Psychiatry 44, 517–525.

2 Lichtermann, D., Hranilovic, D., Trixler, M., Franke, P., Jernej, B., Delmo, C.D., Knapp, M., Schwab, S.G., Maier, W & Wild-enauer, D.B (2000) Support for allelic association of a poly-morphic site in the promoter region of the serotonin transporter gene with risk for alcohol dependence Am J Psychiatry 157, 2045–2047.

3 Blakely, R.D., Berson, H.E., Fremeau, R.T., Caron, M.G., Peek,

M.M., Prince, H.K & Bradley, C.C (1991) Cloning and

expres-sion of a functional serotonin transporter from rat brain Nature

354, 66–70.

4 Hoffman, B.J., Mezey, E & Brownstein, M.J (1991) Cloning of a

serotonin transporter affected by antidepressants Science 254,

579–580.

5 Ramamoorthy, S., Bauman, A.L., Moore, K.R., Han, H.,

Yang-Feng, T., Chang, A.S., Ganapathy, V & Blakely, R.D (1993)

Antidepressant- and cocaine-sensitive human serotonin

transpor-ter: molecular cloning, expression, and chromosomal localization.

Proc Natl Acad Sci USA 90, 2542–2546.

6 Corey, J.L., Quick, M.W., Davidson, N., Lester, H.A &

Guas-tella, J (1994) A cocaine-sensitive Drosophila serotonin

transpor-ter: cloning, expression, and electrophysiological characterization.

Proc Natl Acad Sci USA 91, 1188–1192.

7 Demchyshyn, L.L., Pristupa, Z.B., Sugamori, K.S., Barker, E.L.,

Blakely, R.D., Wolfgang, W.J., Forte, M.A & Niznik, H.B.

(1994) Cloning, expression, and localization of a

chloride-facilitated, cocaine-sensitive serotonin transporter from

Droso-phila melanogaster Proc Natl Acad Sci USA 91, 5158–5162.

8 Chen, J.-G., Liu-Chen, S & Rudnick, G (1997) External cysteine

residues in the serotonin transporter Biochemistry 36, 1479–

1486.

9 Chen, J.G., Sachpatzidis, A & Rudnick, G (1997) The third

transmembrane domain of the serotonin transporter contains

residues associated with substrate and cocaine binding J Biol.

Chem 272, 28321–28327.

10 Barker, E.L., Perlman, M.A., Adkins, E.M., Houlihan, W.J.,

Pristupa, Z.B., Niznik, H.B & Blakely, R.D (1998) High affinity

recognition of serotonin transporter antagonists defined by

spe-cies-scanning mutagenesis: An aromatic residue in transmembrane

domain I dictates species-selective recognition of citalopram and

mazindol J Biol Chem 273, 19459–19468.

11 Mitsuhata, C., Kitayama, S., Morita, K., Vandenbergh, D., Uhl,

G.R & Dohi, T (1998) Tyrosine-533 of rat dopamine transporter:

Involvement in interactions with 1-methyl-4-phenylpyridinium

and cocaine Mol Brain Res 56, 84–88.

12 Sora, I., Wichems, C., Takahashi, N., Li, X.F., Zeng, Z., Revay,

R., Lesch, K.P., Murphy, D.L & Uhl, G.R (1998) Cocaine

reward models: conditioned place preference can be established in

dopamine- and in serotonin-transporter knockout mice Proc.

Natl Acad Sci USA 95, 7699–7704.

13 Itokawa, M., Lin, Z., Cai, N.S., Wu, C., Kitayama, S., Wang, J.B.

& Uhl, G.R (2000) Dopamine transporter transmembrane

domain polar mutants: DeltaG and DeltaDeltaG values implicate

regions important for transporter functions Mol Pharmacol 57,

1093–1103.

14 Lin, Z., Wang, W & Uhl, G.R (2000) Dopamine transporter

tryptophan mutants highlight candidate dopamine- and

cocaine-selective domains Mol Pharmacol 58, 1581–1592.

15 Lin, Z., Itokawa, M & Uhl, G.R (2000) Dopamine transporter

proline mutations influence dopamine uptake, cocaine analog

recognition, and expression, Faseb J 14, 715–728.

16 Rasmussen, S.G., Carroll, F.I., Maresch, M.J., Jensen, A.D., Tate,

C.G & Gether, U (2000) Biophysical characterization of the

cocaine binding pocket in the serotonin transporter using a

fluorescent cocaine analogue as a molecular reporter J Biol.

Chem 276, 4717–4723.

17 Barker, E.L & Blakely, R.D (1996) Identification of a single

amino acid, phenylalanine 586, that is responsible for high affinity

interactions of tricyclic antidepressants with the human serotonin

transporter Mol Pharmacol 50, 957–965.

18 Barker, E.L., Moore, K.R., Rakhshan, F & Blakely, R.D (1999)

Transmembrane domain I contributes to the permeation pathway

for serotonin and ions in the serotonin transporter J

Neuro-science 19, 4705–4717.

19 Osborne, R.H (1996) Insect neurotransmission: Neurotransmit-ters and their receptors Pharmacol Therapeutics 69, 117–142.

20 Mbungu, D., Ross, L.S & Gill, S.S (1995) Cloning, functional expression, and pharmacology of a GABA transporter from Manduca sexta Arch Biochem Biophysics 318, 489–497.

21 Ausubel, F.M (1994) Current Protocols in Molecular Biology John Wiley & Sons, New York.

22 Ross, L.S & Gill, S.S (1996) Limited growth PCR screening of a plasmid library Biotechniques 21, 382–386.

23 Moss, B., Elroy-Stein, O., Mizukami, T., Alexander, W.A & Fuerst, T.R (1990) Product review New mammalian expression vectors Nature 348, 91–92.

24 Sambrook, J., Maniatis, T & Fritsch, E.F (1989) Molecular Cloning: a Laboratory Manual Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.

25 Povlock, S.L & Amara, S.G (1998) Vaccinia virus-T7 RNA polymerase expression system for neurotransmitter transporters Methods Enzymol 296, 436–443.

26 Kozak, M (1986) Point mutations define a sequence flanking the AUG initiator codon that modulates translation by eukaryotic ribosomes Cell 44, 283–292.

27 Ramamoorthy, S., Giovanetti, E., Qian, Y & Blakely, R.D (1998) Phosphorylation and regulation of antidepressant-sensitive serotonin transporters J Biol Chem 273, 2458–2466.

28 Aghazadeh, B & Rosen, M.K (1999) Ligand recognition by SH3 and WW domains: the role of N-alkylation in PPII helices Chem Biol 6, R241–R246.

29 Veyhl, M., Spangenberg, J., Puschel, B., Poppe, R., Dekel, C., Fritzsch, G., Haase, W & Koepsell, H (1993) Cloning of a membrane-associated protein which modifies activity and prop-erties of the Na(+)- D -glucose cotransporter J Biol Chem 268, 25041–25053.

30 Murray, C.L., Quaglia, M., Arnason, J.T & Morris, C.E (1994)

A putative nicotine pump at the metabolic blood–brain barrier of the tobacco hornworm J Neurobiol 25, 23–34.

31 Castagna, M., Shayakul, C., Trotti, D., Sacchi, V.F., Harvey, W.R & Hediger, M.A (1998) Cloning and characterization of a potassium-coupled amino acid transporter Proc Natl Acad Sci USA 95, 5395–5400.

32 Giros, B., el Mestikawy, S., Godinot, N., Zheng, K., Han, H., Yang-Feng, T & Caron, M.G (1992) Cloning, pharmacological characterization, and chromosome assignment of the human dopamine transporter Mol Pharmacol 42, 383–390.

33 Pacholczyk, T., Blakely, R.D & Amara, S.G (1991) Expression cloning of a cocaine- and antidepressant-sensitive human nor-adrenaline transporter Nature 350, 350–354.

34 Burley, S.K & Petsko, G.A (1985) Aromatic–aromatic interac-tion: a mechanism of protein structure stabilization Science 229, 23–28.

35 Nowak, M.W., Kearney, P.C., Sampson, J.R., Saks, M.E., Labarca, C.G., Silverman, S.K., Zhong, W., Thorson, J., Abelson, J.N., Davidson, N et al (1995) Nicotinic receptor binding site probed with unnatural amino acid incorporation in intact cells Science 268, 439–442.

36 Fong, T.M., Yu, H & Strader, C.D (1992) Molecular basis for the species selectivity of the neurokinin-1 receptor antagonists CP-96,345 and Rp67580 J Biol Chem 267, 25668–25671.

37 Horn, A.S (1973) Structure-activity relations for the inhibition of catecholamine uptake into synaptosomes from noradrenaline and dopaminergic neurones in rat brain homogenates British J Pharmacol 47, 332–338.

38 Ritz, M.C., Cone, E.J & Kuhar, M.J (1990) Cocaine inhibition of ligand binding at dopamine, norepinephrine and serotonin trans-porters: a structure-activity study Life Sci 46, 635–645.

39 Giros, B., el Mestikawy, S., Bertrand, L & Caron, M.G (1991) Cloning and functional characterization of a cocaine-sensitive dopamine transporter FEBS Lett 295, 149–154.