Whole spectrum analysis of ligand efficacy at constitutively active human wild-type and S267K 5-HT6 receptors in HEK-293F cells

Abstract

Modulation of constitutive activity by the recombinant wild-type human 5-HT6 receptor was investigated with a series of 5-HT6 receptor ligands by monitoring the cAMP signalling pathway. The impact of the mutation S267K near the B(261)BXXB(265) CIII-loop motif was analyzed on the magnitude of constitutive receptor activity as previously conflicting results have been reported.
The wild-type 5-HT6 receptor plasmid was obtained by PCR and the mutant S267K5-HT6 receptor was constructed by site-directed mutagenesis and stably transfected in HEK-293F cells by electroporation. The cAMP signalling pathway was monitored as a functional read-out to investigate ligands' responses using homogeneous time resolved fluorescence.
Constitutive activity was present both at wild-type and mutant S267K 5-HT6 receptors. Negative efficacy (E(max), % versus basal) as observed at nanomolar concentrations with SB-271046 was larger for mutant (-92+/-1%) than wild-type 5-HT6 receptor (-45+/-1%). Ro 04-6790 also demonstrated negative efficacy at the wild-type 5-HT6 receptor with a magnitude similar to SB-271046 but with a 36-fold lower potency. MS-245 demonstrated at nanomolar concentrations intermediate negative efficacy; -48+/-3% and -16+/-2% at mutant and wild-type 5-HT6 receptor, respectively. The 5-HT-mediated cAMP response was blocked by SB-271046, MS-245 and Ro 04-6790 to their respective level of negative efficacy with pKB values fitting with their binding pK(i) values. E-6801 was a highly potent (pEC50: 10.17 to 10.19) and efficacious agonist (+98 to +102% versus 5-HT) at both wild-type and mutant 5-HT6 receptors.
The recombinant wild-type human 5-HT6 receptor is constitutively active in HEK-293F cells and displays a high resolution to monitor efficacy properties of 5-HT6 receptor ligands. The resolution capacity to differentiate between efficacy properties of 5-HT6 receptor ligands, in particular for negative efficacy, can be further enhanced by monitoring the mutant S267K 5-HT6 receptor.

Full text

Original article
Whole spectrum analysis of ligand efficacy at constitutively active human
wild-type and S267K 5-HT6 receptors in HEK-293F cells
Gonzalo Romero, Marta Pujol, Pilar Pérez, Helmut Buschmann, Petrus J. Pauwels ⁎
Laboratorios Dr. Esteve S.A., Av. Mare de Déu de Montserrat 221, 08041 Barcelona, Spain
Received 24 March 2006; accepted 30 April 2006
Abstract
Introduction: Modulation of constitutive activity by the recombinant wild-type human 5-HT6 receptor was investigated with a series of 5-
HT6 receptor ligands by monitoring the cAMP signalling pathway. The impact of the mutation S267K near the B
261
BXXB
265
CIII-loop motif was
analyzed on the magnitude of constitutive receptor activity as previously conflicting results have been reported. Methods: The wild-type 5-HT6
receptor plasmid was obtained by PCR and the mutant S267K5-HT6 receptor was constructed by site-directed mutagenesis and stably transfected
in HEK-293F cells by electroporation. The cAMP signalling pathway was monitored as a functional read-out to investigate ligands' responses
using homogeneous time resolved fluorescence. Results: Constitutive activity was present both at wild-type and mutant S267K 5-HT6 receptors.
Negative efficacy (E
max
,%versus basal) as observed at nanomolar concentrations with SB-271046 was larger for mutant (−92 ± 1%) than wild-
type 5-HT6 receptor (−45 ± 1%). Ro 04-6790 also demonstrated negative efficacy at the wild-type 5-HT6 receptor with a magnitude similar to SB-
271046 but with a 36-fold lower potency. MS-245 demonstrated at nanomolar concentrations intermediate negative efficacy; −48 ± 3% and −16 ± 2%
at mutant and wild-type 5-HT6 receptor, respectively. The 5-HT-mediated cAMP response was blocked by SB-271046, MS-245 and Ro 04-6790 to
their respective level of negative efficacy with pKB values fitting with their binding pK
i
values. E-6801 was a highly potent (pEC50: 10.17 to 10.19)
and efficacious agonist (+98 to +102% versus 5-HT) at both wild-type and mutant 5-HT6 receptors. Discussion: The recombinant wild-type human
5-HT6 receptor is constitutively active in HEK-293F cells and displays a high resolution to monitor efficacy properties of 5-HT6 receptor ligands.
The resolution capacity to differentiate between efficacy properties of 5-HT6 receptor ligands, in particular for negative efficacy, can be further
enhanced by monitoring the mutant S267K 5-HT6 receptor.
© 2006 Elsevier Inc. All rights reserved.
Keywords: Adenylyl cyclase; Constitutive activation; Cyclic-AMP (cAMP); HEK-293F cell line; Human 5-HT6 receptor; Ligand efficacy; Methods; Mutagenesis
1. Introduction
The 5-hydroxytryptamine6(5-HT6) receptor is one of the most
recent additions to the 5-HT receptor family. Selective 5-HT6
receptor antagonists have recently been developed and potential
functional roles are now becoming apparent (Holenz et al., 2006;
Mitchell & Neumaier, 2005; Reavill & Rogers, 2001; Woolley,
Marsden, & Fone, 2004). The high affinity of a wide range of
psychiatric drugs for the 5-HT6 receptor, together with its almost
exclusive expression in the central nervous system, being
abundant in limbic and cortical regions, has stimulated significant
research interest. The 5-HT6 receptor appears to regulate
glutamatergic and cholinergic neuronal activity (Wool ley et a l.,
2004). Increasing evidence suggests that it may be involved in the
regulation of cognition, feeding and possibly, affective states and
seizures (Holenz et al., 2006). Despite significant sequence
homology between human, rat and mouse 5-HT6 receptors, the
central nervous system distribution and pharmacological profile
of mouse 5-HT6 receptor is significantly different from the rat and
human 5-HT6 receptors (Hirst et al., 2003). The 5-HT6 receptor
belongs to the G protein-coupled receptor family and is coupled to
the Gs-family of G proteins and has been demonstrated to increase
cAMP formation in recombinant expression systems (Boess et al.,
Journal of Pharmacological and Toxicological Methods 55 (2007) 144 –150
www.elsevier.com/locate/jpharmtox
Abbreviations: 5-HT, 5-hydroxytryptamine; E-6801, 6-chloro-N-(3-(2-
(dimethylamino)ethyl)-1H-indol-5-yl)imidazo[2,1-b]thiazole-5-sulfonamide;
HEK, human embryonic kidney; MS-245, 2-(5-methoxy-1-(phenylsulfonyl)-
1H-indol-3-yl)-N,N-dimethylethanamine; Ro 04-6790, N-(2,6-bis(methylamino)
pyrimidin-4-yl)-4-aminobenzenesulfonamide; SB-271046, 5-chloro-N-(4-meth-
oxy-3-(piperazin-1-yl)phenyl)-3-methylbenzo[b]thiophene-2-sulfonamide.
⁎Corresponding author. Tel.: +34 934 46 60 63.
E-mail address: ppauwels@esteve.es (P.J. Pauwels).
1056-8719/$ - see front matter © 2006 Elsevier Inc. All rights reserved.
doi:10.1016/j.vascn.2006.04.007
72364

1997; Kohen et al., 1996; Ruat et al., 1993) besides cultured
mouse striatal neurones (Sebben, Ansana, Bockaert, & Dumuis,
1994) and pig caudate membranes (Schoeffter & Waeber, 1994).
The wild-type mouse 5-HT6 receptor has been reported to
yield strong constitutive activity when expressed at low levels in
JEG-3 and Cos-7 cells (Kohen, Fashingbauer, Heidmann,
Guthrie, & Hamblin, 2001). Otherwise, the native from of the
human 5-HT6 receptor did neither display constitutive activity in
Cos-7 nor CHO-K1 cells (Purohit, Herrick-Davis, & Teitler,
2003; Purohit, Smith, Herrick-Davis, & Teitler, 2005). However,
mutation of the Ser267 to a Lys near the B
261
BXXB
265
CIII-loop
motif of the human 5-HT6 receptor yielded efficiently robust
agonist-independent activity when expressed both in Cos-7 and
CHO-K1 cells (Purohit et al., 2003, 2005). These notwithstanding
alterations in the same BBXXB CIII-loop motif of the murine 5-
HT6 receptor reduce rather than further activate basal activity
(Kohen et al., 2001). Hence, conflicting results on mouse and
human 5-HT6 receptor constitutive activity have been obtained.
In particular, the issue of wild-type versus mutant 5-HT6 receptor
constitutive activity and the contribution of mutations in the area
of its BBXXB CIII-loop motif are contradicting.
In the present paper, we report on a large magnitude of
constitutive activity by the wild-type human 5-HT6 receptor
when stably expressed in human embryonic kidney (HEK)-293F
cells. In addition, we could further amplify this magnitude of 5-
HT6 receptor-mediated constitutive activity by mutation of its
Ser267 near the B
261
BXXB
265
CIII-loop motif to a Lys. As little
is known about most 5-HT6 receptor antagonists with regard to
their intrinsic efficacy properties (see Davies, Silvestre, & Guitart,
2005), their degree and type of efficacy may ultimately
differentiate these compounds at the 5-HT6 receptor, in particular
when it is constitutively active or under the tonic control of 5-HT
(Woolley et al., 2004). We report in this study on several 5-HT6
compounds which could be differentiated on the basis of their
intrinsic activity at the constitutively active 5-HT6 receptor. The
selection contained the recently reported (Holenz et al., 2005)3-
aminoalkylindolyl sulphonamide derivative: E-6801 (6-chloro-N-
(3-(2-(dimethylamino)ethyl)-1H-indol-5-yl)imidazo[2,1-b]thia-
zole-5-sulfonamide) as a partial 5-HT6 receptor agonist, and the
presumably putative 5-HT6 receptor antagonists SB-271046
(Bromidge et al., 1999), Ro 04-6790 (Sleight et al., 1998)and
MS-245 (Abate et al., 2005) with a cAMP signalling pathway as a
functional read-out. It appears that the herein described 5-HT6
receptor expression systems are useful to differentiate between the
efficacies of the investigated 5-HT6 compounds.
2. Methods
2.1. Construction of human 5-HT6 receptor plasmid
PCR was performed with cDNA from human brain as a template
using Expam High Fidelity polymerase (Roche). Flanking primers
for human 5-HT6 receptor were 5′-CCTCGGTC-CTCATGGTCC-
CAG-3′and 5′-CCAAGCCCGGGTCAGTT-CGTG-3′. After
sequential PCR, products were visualized on 0.8% agarose gel
and visualized with ethidium bromide (10 μg/ml) staining. PCR
products were purified using Microcon PCR columns (Millipore)
following the manufacturer's instructions. The purified fragments
were ligated into pCR3.1 vector (Invitrogen) usingT4 DNA ligase
and after overnight incubation at 15 °C, ligated fragments were
introduced in competent Escherichia coli TOP-F10 cells.
Fifty microliters of cells were mixed with 2 μl of ligation product
and set in ice for 30 min. The mixture was incubated for 30 s at
42 °C (heat shock) and replaced at 37 °C for 1 h with additional
250 μl sodium bactotryptone (SOC) medium. Afterwards, cells
were seeded into Luria-Bertani (LB)-agar medium with ampicillin
(100 μg/ml) and the culture was grown at 37 °C overnight. Single
colonies were selected and put into 3 ml LB with 100 μg/ml
ampicillin and incubated at 37 °C overnight. cDNA was obtained
using a Qiaprep Spin Miniprep kit (Qiagen). Plasmids from se-
lected colonies were sequenced and one corresponding to the ori-
ginal human 5-HT6 receptor DNA sequence (Kohen et al., 1996)
was taken for stable expression in HEK-293F cells.
2.2. Construction of human mutant S267K 5-HT6 receptor
plasmid
The S267K mutation was performed on the native human 5-
HT6 receptor cDNA (Kohen et al., 1996) and constructed by site-
directed mutagenesis (Purohitetal.,2003). Two flanking primers
(sense 5′-GGAAGGCCCTGAAGGCCAAGCTTACGCTGGG-
CATCCTGC-3′and antisense 5′-GCAGGATGCCCAGCGTA-
AGCTTGGCCTTCAGGGCCTTCC-3′) were designed for muta-
genesis of serine (AGC)-267 to lysine (AAG). PCR was performed
using Pfu Turbo DNA Polymerase enzyme and the pCR3.1-mutant
5-HT6 fragment was transformed in DH5αcells. A HindIII
restriction site was added by changing the wobble base of the
adjacent leucine from CTG to CTT in order to verify its orientation
and the sequence was confirmed by DNA sequencing.
2.3. Construction of HEK-293F cell lines stably expressing
either wild-type or mutant S267K 5-HT6 receptor
HEK-293F cells (Gibco) were grown in Dulbecco's Modified
Eagle's Medium with GlutaMAX and pyruvate (DMEM, Gibco),
and supplemented with 10% foetal bovine serum (Gibco), peni-
cillin (50 U/ml) and streptomycin (50 U/ml) (Gibco). Cells were
electroporated (950 μF/250 V, GenePulserII, Biorad) with
pcDNA3.1 containing respective wild-type or mutant 5-HT6 re-
ceptor plasmid (15 μg). Forty-eight hours after transfection, cells
were seeded by serial dilutions and plated in 384-well plates con-
taining G418 (geneticin, Gibco) at 0.5 mg/ml. Isolated single colo-
nies of cells to the G418-resistant phenotype were expanded and
assayed for their 5-HT-mediated cAMP response using homoge-
neous time resolved fluorescence (HTRF). One clone for either
wild-type human 5-HT6 receptor, #10P, and mutant 5-HT6 receptor,
#3, were selected. Stably transfected cells were always grown in the
presence of 0.5 mg/ml G-418 except during the cAMP experiments.
2.4. Transient transfection of human wild-type and mutant
S267K 5-HT6 receptor in Cos-7 cells
Cos-7 cells were cultured in DMEM supplemented with
40 mM glutamine and 10% of foetal bovine serum. Twenty-four
145G. Romero et al. / Journal of Pharmacological and Toxicological Methods 55 (2007) 144–150

hours before transfection, cells were seeded at subconfluent state
in 150-mm Petri dishes for cAMP assay by HTRF. Cells were
transfected with pCR3.1-vector containing either wild-type or
mutant h5-HT6 receptor cDNA using 150 μl of lipofectamine and
37.5 μg of plasmid DNA per plate. cAMP experiments were
performed 48 h after transfection.
2.5. Radioligand binding assay
The expression of 5-HT6 receptor was measured by
radioligand binding assay (Hirst et al., 2003) in a 96-well plate
with a total reaction volume of 200 μl, containing 100 μlof
membrane suspension (25 μg protein/well), 10 μlof[
3
H]-LSD
(2.5 to 10.0 nM) in either absence or presence of 90 μlofeither
buffer or methiothepin (5 μM) for total and non-specific binding,
respectively. Binding buffer contained 50 mM Tris–HCl, 10 mM
MgCl2 and 0.5 mM EDTA at pH 7.4. Plates were incubated at
37 °C for 60 min, filtered and plates were washed 3 times with ice-
cold 50 mM Tris–HCL (pH 7.4). Filters were dried and counted at
approximately 40% efficiency in a MicroBeta scintillation
counter (Perkin-Elmer) using 25 μl per well of EcoScint liquid
scintillation cocktail. To investigate binding properties of 5-HT6
receptor ligands to h5-HT6 receptor, transfected HEK-293
membranes (35 μg protein/assay) from Perkin-Elmer (Boston,
MA, USA) and 2.5 nM [
3
H]-LSD were used.
2.6. Measurement of cAMP responses by homogeneous time
resolved fluorescence
After overnight serum-free medium incubation, cAMP mea-
surements on either Cos-7 or HEK-293F cells that expressed 5-
HT6 receptor were performed using HTRF (Gabriel et al., 2003).
Cell suspension (20,000 cells per well) was added in 96-well
culture plate in incubation buffer composed of Ham's F12
medium plus 1 mM 3-isobutyl-1-methyl-xanthine (IBMX) and
20 μM pargyline. Forty microliters of cell suspension and 10 μlof
either compound or vehicle were added to each well at indicated
concentrations for 30 min at 37 °C, in either absence or presence
(in antagonist experiments) of 5-HT. The reaction was stopped
with 25 μlofcryptateand25μl of cross-linked allophycocyanin
(XL-665). Plates were incubated for 1 h at room temperature and
read at 665 nm/620 nm using a RubyStar Plate reader (BMG
LabTech).
2.7. Materials
pCR3.1 plasmid, Lipofectamine2000 and other reagents for
molecular biology experiments were purchased from either
Invitrogen (Frederick, MD, USA), Qiagen (Germantown, M.D.
USA) or Roche (Penzberg, Germany) as indicated above. Site-
directed mutagenesis kit was obtained from Stratagene ( La Jolla,
CA, USA). Cell culture media and reagents were purchased from
Gibco (Paislay, UK). HTRF cAMP kit was purchased from
CisBio (Bagnols, France). [
3
H]-LSD was purchased from NEN
(Boston, MA, USA). 5-Hydroxytryptamine, dimethyl sulphoxide
(DMSO), 3-isobutyl-1-methyl-xanthine (IBMX) and pargyline
were obtained from Sigma (Poole, UK). Methiothepin was
obtained from Tocris (Bristol, UK). SB-271046, Ro 04-0670 and
MS-245 were prepared intramuros. E-6801 is described in WO
2003/042175 A1 (Merce-Vidal et al., 2003). Stock compound
solutions were prepared in DMSO and diluted with phosphate
buffer solution (PBS) not exceeding 2% of DMSO at final
concentration.
2.8. Data analysis
cAMP data are reported as mean± S.E.M. of at least six
independent experiments, each of which was performed in
duplicate. Data were systematically transformed to pmol/10
6
cells from a standard curve using cAMP standard solution. Per-
centage of cAMP formed was calculated versus basal levels. The
response either to modulate basal cAMP formation (E
max
)by
compounds was determined from the maximal stimulation or inhi-
bition value that corresponded to a plateau value. The concentration
of compound that produced a half-maximal response is represented
by a pEC50 value and was calculated as nonlinear regression
curves using XLfit (IDBS) and GraphPad Prism Version 4 prog-
rams. Either one-way or two-way ANOVA statistical analyses was
performed using SAS program (SAS Institute Inc., Cary, NC,
USA). Antagonist potency against 5-HT was expressed as pKB
values and is −log of antagonist equilibrium dissociation constant,
i.e., −log (antagonist concentration/(concentration ratio−1))
Table 1
cAMP and receptor expression levels of either wild-type or mutant S267K 5-HT6 receptor in Cos-7 and HEK-293F cells
Cell type Cos-7 HEK-293F
h5-HT6 receptor Wild-type Ser267Lys Wild-type Ser267Lys
cAMP formation (pmol/10
6
cells)
Basal 6.82 ± 0.25 10.18 ± 0.38
a
4.59 ± 0.18 13.69 ± 0.34
a
1μM 5-HT 19.09 ± 0.66 20.13 ± 1.09 21.76 ± 0.67 21.78 ± 0.23
1μM 5-HT + 1 μM SB-271046 6.63 ± 0.28 8.27 ± 0.48 4.43 ± 0.42 6.35 ± 0.62
1μM SB-271046 6.61 ± 0.38 6.43 ± 0.25
b
2.44 ± 0.21
c
1.28 ± 0.25
b
3
H-LSD binding (pmol/mg protein) 1.08 ± 0.03 0.66 ± 0.01 4.32 ± 0.39 9.61 ± 0.21
cAMP and receptor expression levels were measured as described in Methods. Data correspond to mean ± S.E.M. values of ten to sixteen independent experiments
performed in duplicate.
a
pb0.001 versus basal value of wild-type 5-HT6 receptor.
b
pb0.001 versus respective basal value of mutant 5-HT6 receptor.
c
pb0.01 versus basal value of wild-type 5-HT6 receptor in HEK-293F cells.
146 G. Romero et al. / Journal of Pharmacological and Toxicological Methods 55 (2007) 144–150

where concentration ratio is the ratio of agonist EC50 in absence
and presence of antagonist (Furchgott, 1966).
3. Results
3.1. Comparison between constitutive activity of wild-type and
mutant S267K 5-HT6 receptors in Cos-7 and HEK-293F cells
A comparison between expression and cAMP levels of wild-
type and mutant S267K 5-HT6 receptors in either Cos-7 or HEK-
293F cells is summarized in Tab le 1. Transient expression of wild-
type 5-HT6 receptor in Cos-7 cells yielded 1 pmol of [
3
H]-LSD
binding sites/mg protein. 5-HT (1 μM) stimulated cAMP
formation by a factor 2.8. The putative 5-HT6 receptor antagonist
SB-271046 (1 μM) fully blocked the 5-HT response without
affecting basal cAMP formation. Transient expression of the
mutant S267K 5-HT6 receptor in Cos-7 cells demonstrated about
40% less [
3
H]-LSD binding sites/mg protein. Basal cAMP
formation was enhanced by 49% and was fully blocked by SB-
271046 (1 μM). The magnitude of the 5-HT response was similar
to the wild-type 5-HT6 receptor and also blocked by SB-271046
(1 μM). Stable expression of either wild-type 5-HT6 or S267K 5-
HT6 receptor inHEK-293F cells yielded higher expression levels
of [
3
H]-LSD binding sites/mg protein. Basal cAMP formation
was strongly enhanced, in particular for the mutant S267K 5-HT6
receptor and attenuated by SB-271046 (1 μM). The magnitude of
the 5-HT response was similar for both wild-type and mutant 5-
Fig. 1. Modulation of cAMP levels by 5-HT6 receptor ligands at wild-type 5-
HT6 receptor in stably transfected HEK-293F cells. HEK-293F cells were stably
transfected with wild-type 5-HT6 receptor as described in Methods. Upon
overnight serum-free incubation, cAMP formation was determined in the
presence of indicated concentrations of ligands using HTRF. cAMP formation is
expressed as pmol/10
6
cells, whereas Insert illustrates inverse agonist responses
expressed as percentage of the respective basal cAMP level. Dose–response
curves are shown from six to twelve independent experiments performed in
duplicate. Mean E
max
and pEC50 values±S.E.M. are summarized in Table 2.
Symbols: ▴E-6801, ▪5-HT, ▾MS-245, ♦Ro 04-6790, ●SB-271046.
Table 2
E
max
, pEC50 and pKB values of several 5-HT6 ligands for modulating cAMP formation in HEK-293F cells stably expressing either wild-type or mutant 5-HT6
receptor, and corresponding pK
i
values
h5-HT6 receptor cAMP formation h5-HT6 binding
Wild-type Ser267Lys Wild-type
Compound E
max
(%) pEC50 pKB E
max
(%) pEC50 pKB pK
i
5-HT 394.8± 25.6 9.02 ± 0.09
a
–62.1 ± 6.4 9.08 ± 0.10
a
–6.96 ± 0.18
E-6801 388.4 ± 42.6 10.19 ± 0.14
a
–63.3 ± 6.6 10.17 ± 0.18
a
–8.51 ± 0.15
MS-245 −16.2 ± 1.8
b
8.08 ± 0.10 7.94 ± 0.10 −48.4 ± 2.6
c,d
7.84 ± 0.12 7.70 ± 0.18 7.91 ± 0.02
Ro 04-6790 −40.8 ± 1.9 6.80 ± 0.09 6.55 ± 0.13 −69.1 ± 1.6
e
6.77 ± 0.06 6.45 ± 0.12 6.91 ± 0.05
SB-271046 −45.0 ± 1.3 8.36 ± 0.13
f
9.08 ± 0.09 −92.3 ± 1.2
b,g
8.05 ± 0.06
h
8.70 ± 0.13 8.68 ± 0.09
E
max
(% versus basal), pKB and pEC50 values were derived from ligand-mediated cAMP curves as illustrated in Figs. 1–4. Data correspond to mean ± S.E.M. values
for a minimum of six independent experiments performed in duplicate. pK
i
values for wild-type 5-HT6 receptor were obtained as described in Methods. a: pb0.001
versus its corresponding pK
i
value; b: pb0.001 and c: pb0.05 versus E
max
value of Ro 04-6794; d: pb0.01, e: pb0.05, and g: pb0.001 versus its corresponding E
max
value at wild-type 5-HT6 receptor; f: pb0.05 and h: pb0.01 versus its corresponding pK
i
value. pKB values are not significantly different from pK
i
values.
Fig. 2. cAMP levels upon co-incubation of 5-HTwith 5-HT6 receptor ligands at wild-
type 5-HT6 receptor in stably transfected HEK-293F cells. HEK-293F cells were
stably transfected with wild-type 5-HT6 receptor as described in Methods. Upon
overnight serum-free incubation, increasing concentrations of 5-HT were combined
either without or with a fixed concentration (1 μM) of 5-HT6 receptor ligand in order
to monitor cAMP formation using HTRF. cAMP formation is expressed as a
percentage versus basal cAMP levels. Dose–response curves are shown from six to
twelve independent experiments performed in duplicate. Symbols: ▴E-6801, ▪
vehicle, ♦Ro 04-6790, ▾MS-245, ●SB-271046.
147G. Romero et al. / Journal of Pharmacological and Toxicological Methods 55 (2007) 144–150

HT6 receptors and fully antagonized by SB-271046 (1 μM).
Further experiments were performed with the HEK-293F cell
lines stably expressing either wild-type or S267K 5-HT6 receptor
as they demonstrated a larger magnitude of constitutive activity as
compared to Cos-7 cells.
3.2. Comparison between 5-HT6 receptor ligand-mediated
cAMP responses at wild-type 5-HT6 receptor in HEK-293F
cells
Fig. 1 illustrates cAMP responses as mediated by several 5-
HT6 receptor ligands at wild-type 5-HT6 receptor in stably
transfected HEK-293F cells. The 5-HT-mediated cAMP re-
sponse was potent (pEC50: 9.02) and at a 116-fold lower con-
centration than its binding affinity (Table 2). E-6801 was as
efficacious as 5-HT but 15-fold more potent. MS-245 demon-
strated some partial inverse agonist activity (E
max
,%versus
basal: −16 ± 2%) as compared to Ro 04-6790 and SB-247016
(E
max
:−41 ± 2% and −45 ± 1%, respectively) at concentrations
in accordance with their pK
i
values. Fig. 2 shows 5-HT-mediated
cAMP responses in either absence or presence of 1 μM of the
herein investigated 5-HT6 receptor ligands. The combination of
E-6801 and 5-HT maintained maximal cAMP formation,
whereas rightward shifts of the 5-HT dose response curve
were obtained with SB-271046, MS-245 and Ro 04-6790 to
their respective negative efficacy values in the absence of 5-HT.
Antagonist potency of SB-271046 was, respectively, 14- and
340-fold higher than for MS-245 and Ro 04-6790(Fig. 2).
3.3. Comparison between 5-HT6 receptor ligand-mediated
cAMP responses at mutant S267K 5-HT6 receptor in
HEK-293F cells
Fig. 3 illustrates cAMP responses as mediated by several 5-
HT6 receptor ligands at mutant S267K 5-HT6 receptor in stably
transfected HEK-293F cells. Table 2 compares E
max
and pEC50
values for these ligands as obtained with wild-type and mutant
5-HT6 receptors. 5-HT and E-6801 were equipotent at both
wild-type and mutant S267K 5-HT6 receptors. MS-245 be-
haved as a more efficacious partial inverse agonist at the S267K
5-HT6 receptor with a pEC50 value similar to its pK
i
value. The
amplitude of negative efficacy by SB-271046 (E
max
, % versus
basal. −92 ± 1) was larger than that observed with Ro 04-6790
(E
max
:−69 ± 2%) while their inverse agonist potencies were not
affected. Similar to wild-type 5-HT6 receptor, the combination
of E-6801 (1 μM) and 5-HT at the mutant S267K 5-HT6
receptor maintained maximal cAMP formation, whereas SB-
271046 (1 μM) induced a large rightward shift accompanied
with a strong decrease in basal activity (Fig. 4). MS-245 and Ro
04-6790 demonstrated a similar pattern and attenuated basal
activity to their respective negative efficacy level in the absence
of 5-HT. The antagonist potency of Ro 04-6790 was weakest as
compared to SB-271046 and MS-245 (Fig. 4).
4. Discussion
The present paper reports on the constitutive activity by the
human wild-type and mutant S267K 5-HT6 receptor in HEK-293F
cells and their response to several 5-HT6 receptor ligands. The
amplitude of agonist-independent activity was large (45%) at the
Fig. 3. Modulation of cAMP levels by 5-HT6 receptor ligands at mutant S267K 5-
HT6 receptors in stably transfected HEK-293F cells. HEK-293F cells were stably
transfected with S267K 5-HT6 receptor as described in Methods. Upon overnight
serum-free incubation, cAMP formation was determined in the presence of indicated
concentrations of ligands using HTRF. cAMP formation is expressed as pmol/10
6
cells, whereas Insert illustrates inverse agonist responses expressed as percentage of
the respective basal cAMP level. Dose–response curves are shown from six to
twelve independent experiments performed in duplicate. Mean E
max
and pEC50
values ± S.E.M. are summarized in Tab le 2 . Symbols: ▴E-6801, ▪5-HT, ▾MS-
245, ♦Ro 04-6790, ●SB-271046.
Fig. 4. cAMP levels upon co-incubation of 5-HT with 5-HT6 receptor ligands at
mutant S267K 5-HT6 receptor in stably transfected HEK-293F cells. HEK-293F
cells were stably transfected with S267K 5-HT6 receptor as described in Methods.
Upon overnight serum-free incubation, increasing concentrations of 5-HT were
combined either without or with a fixed concentration (1 μM) of 5-HT6 receptor
ligand in order to monitor cAMP formation using HTRF. cAMP formation is
expressed as a percentage of the respectivebasal cAMP level. Dose–response curves
are shown from six to twelve independent experiments performed in duplicate.
Symbols: ▴E-6801, ▪vehicle, ♦Ro 04-6790, ▾MS-245, ●SB-271046.
148 G. Romero et al. / Journal of Pharmacological and Toxicological Methods 55 (2007) 144–150

wild-type 5-HT6 receptor. It could be further enhanced to 92% by
mutation of Ser267 to a Lys near its B
261
BXXB
265
CIII-loop motif.
The 5-HT receptor ligands investigated herein displayed a similar
qualitative response at the wild-type and mutant 5-HT6 receptor
while some quantitative differences were observed. The mutant
S267K 5-HT6 receptor has previously been shown to be
constitutively active (Purohit et al., 2003, 2005); however, these
authors could not observe constitutive activity by the wild-type 5-
HT6 receptor in either Cos-7 or CHO-K1 cells. Reported 5-HT6
receptor expression levels (6–8 pmol/mg protein) were similar to
our HEK-293F expression system; however, different host cell
types CHO and Cos-7 were used instead of HEK-293F cells. This
also fits with the lack of inverse agonist activity by SB-271046 at
the wild-type 5-HT6 receptor in Cos-7 cells in the present study. In
contrast to the mouse 5-HT6 receptor where alterations in the same
BBXXB CIII-loop motif reduce rather than further activate basal
activity (Kohen et al., 2001), the human 5-HT6 receptor-mediated
constitutive activity is enhanced by the mutation of Ser267 to Lys
near the B
261
BXXB
265
CIII-loop motif in agreement with several
reports on a series of different receptor subtypes (see Pauwels &
Wurch, 1998).
Constitutively active recombinant expression systems are
well known for their capacity to differentiate between closely
related compounds with respect to their intrinsic efficacy
properties (i.e., Pauwels, Tardif, Wurch, & Colpaert, 2000).
Though there is no doubt that these model systems are useful
to differentiate between functional properties of compounds,
some caution should be taken to extrapolate findings from such
recombinant systems to in vivo integrated systems. Indeed, in
the absence of constitutive activity, inverse agonists behave as
competitive antagonists (see Kenakin, 2004). The herein
described 5-HT6 receptor experimental systems have taken
advantage of monitoring the native Gs-coupled cAMP
pathway. Besides a valuable sensitivity to inverse agonism,
high sensitivity to agonist features of 5-HT6 compounds was
also observed as with E-6801. Indeed, E-6801 was herein a
potent and efficacious agonist at both wild-type and mutant 5-
HT6 receptors. Its E
max
value suggests similar efficacy to 5-
HT.
SB-271046 behaved as a 5-HT6 receptor antagonist against
5-HT in accordance with the report of Routledge et al. (2000).
This compound was reported as virtually free of intrinsic
activity at the human 5-HT6 receptor but it was presumably
measured under silent 5-HT6 receptor conditions. The present
study demonstrates SB-271046 displays negative efficacy at
nanomolar concentrations at both wild-type and mutant S267K
5-HT6 receptors. We do not exclude that other compounds
may achieve a larger magnitude of negative efficacy. Negative
efficacy has been previously reported at a constitutively active
S267K 5-HT6 receptor for the atypical antipsychotic clozapine
(Purohit et al., 2003; Purohit et al., 2005; Teitler, Herrick-
Davis, & Purohit, 2002) and typical antipsychotic fluphenazine
(Purohit et al., 2005). We suggest SB-271046 is a 5-HT6
receptor inverse agonist/antagonist and the inverse agonist
property, though its physiological role is not well-defined (see
Kenakin, 2004), may be of importance under both acute and
chronic constitutively active 5-HT6 receptor conditions. In
these instances in which negative efficacy is expressed, there
may be conditions in which this is a useful property (i.e.,
reduce pathologically induced constitutive activity) or an
undesired property (tolerance to antagonism). Ro 04-6790 also
behaved as a 5-HT6 receptor inverse agonist and antagonist of
5-HT. Sleight et al. (1998) reported this compound had no
effect on basal cAMP accumulation in HeLa cells stably
expressing human 5-HT6 receptor, suggesting Ro 04-6790 is
neither agonist nor inverse agonist. The present study supports
Ro 04-6790 is an inverse agonist/antagonist at the wild-type 5-
HT6 receptor. We observed Ro 04-6790 displayed partial
inverse agonism as compared to SB-271046 at the mutant
S267K 5-HT6 receptor. This also suggests Ro 04-6790 and
SB-271046 are differentially affected by the S267K mutation.
Moreover, both compounds also interact in a different way
with the mouse 5-HT6 receptor (Hirst et al., 2003); Ro 04-
6790 displayed a 1900-fold decrease in affinity at the mouse
receptor whereas SB-271046 was not affected. MS-245
displayed antagonism of the 5-HT response in agreement
with the observations of Abate et al. (2005). This compound
illustrated partial negative efficacy (−36% as compared to SB-
271046) at the wild-type 5-HT6 receptor; its magnitude of
negative efficacy was amplified to −52% at the mutant S267K
5-HT6 receptor which further suggests that this compound is not
silent, neither at the wild-type nor mutant 5-HT6 receptor.
However, this compound demonstrates clearly less negative
efficacy than SB-271046 and Ro 04-6790.
It also seems that each of the herein investigated 5-HT6
compounds actually possess intrinsic activity. However, the
“observable”magnitude of this effect is likely to be variable
and dependent on the experimental model system. Non-
observance of efficacy does not necessarily imply absence of
efficacy. The experimental conditions must be appropriate for
the effect to be monitored. The challenge is still to find a
neutral, silent 5-HT6 receptor antagonist in order to learn more
about the advantages/disadvantages under physiological and
pathological CNS conditions of either a neutral antagonist
versus either a partial agonist or inverse agonist. The herein
described constitutively active 5-HT6 receptor model systems
will be further useful to identify truly neutral, silent 5-HT6
receptor antagonists as they can exclude any ligand with
efficacy.
In conclusion, the constitutively active wild-type human 5-
HT6 receptor expression system HEK-293F displays a high
resolution to monitor efficacy properties of 5-HT6 compounds.
Besides SB-271046 and Ro 04-6790 which illustrate an
inverse agonist/antagonist phenotype, MS-245 is a partial
inverse agonist and E-6801 is a potent and efficacious agonist
at the human wild-type 5-HT6 receptor. The resolution
capacity to differentiate between efficacy properties of 5-
HT6 receptor ligands, in particular for negative efficacy, can be
further enhanced by monitoring the mutant S267K 5-HT6
receptor.
Acknowledgements
We sincerely thank A. Dordal for providing pK
i
values.
149G. Romero et al. / Journal of Pharmacological and Toxicological Methods 55 (2007) 144–150

References
Abate, C., Kolanos, R., Dukat, M., Setola, V., Roth, B. L., & Glennon, R. A.
(2005). Interaction of chiral MS-245 analogs at h5-HT6 receptors.
Bioorganic and Medicinal Chemistry Letters,15, 3510−3513.
Boess, F. G., Monsma, F. J., Carolo, C., Meyer, V., Rudler, A., Zwingelstein, C.,
et al. (1997). Functional and radioligand binding characterization of rat 5-
HT6 receptors stably expressed in HEK293 cells. Neuropharmacology,36,
713−720.
Bromidge, S. M., Brown, A. M., Clarke, S. E., Dodgson, K., Gager, T., Grassam,
H. L., et al. (1999). 5-Chloro-N-(4-methoxy-3-piperazin-1-yl- phenyl)-3-
methyl-2-benzothiophenesulfon-amide (SB-271046): A potent, selective,
and orally bioavailable 5-HT6 receptor antagonist. Journal of Medicinal
Chemistry,42, 202−205.
Davies, S. L., Silvestre, J. S., & Guitart, X. (2005). Drug discovery targets: 5-HT
6
receptor. Drugs of the Future,30, 479−495.
Furchgott, R. F. (1966). The use of b-haloalkylamines in the differentiation of
receptors and in the determination of dissociation constants of receptor-
agonist complexes. Advances in Drug Research,3,21−35.
Gabriel, D., Vernier, M., Pfeifer, M. J., Dasen, B., Tenaillon, L., & Bouhelal, R.
(2003). High throughput screening technologies for direct cyclic AMP
measurement. Assay in Drug Development Technology,1, 291−303.
Hirst, W. D., Abrahamsen, B., Blaney, F. E., Calver, A. R., Aloj, L., Price, G. H.,
et al. (2003). Differences in the central nervous system distribution and
pharmacology of the mouse 5-hydroxytryptamine-6 receptor compared with
rat and human receptors investigated by radioligand binding, site-directed
mutagenesis, and molecular modeling. Molecular Pharmacology,64,
1295−1308.
Holenz, J., Merce, R., Diaz, J. L., Guitart, X., Codony, X., Dordal, A., et al.
(2005). Medicinal chemistry driven approaches toward novel and selective
serotonin 5-HT6 receptor ligands. Journal of Medicinal Chemistry,48,
1781−1795.
Holenz, J., Pauwels, P. J., Diaz, J. L., Mercé, R., Codony, X., & Buschmann, H.
(2006). Recent development of 5-HT6 ligands as potential cognitive
enhancers and anti-obesity compounds. Drug Discovery Today,11,
283−299.
Kenakin, T. (2004). Efficacy as a vector: The relative prevalence and paucity of
inverse agonism. Molecular Pharmacology,65,2−11.
Kohen, R., Fashingbauer, L. A., Heidmann, D. E., Guthrie, C. R., & Hamblin,
M. W. (2001). Cloning of the mouse 5-HT6 serotonin receptor and
mutagenesis studies of the third cytoplasmic loop. Brain Research.
Molecular Brain Research,90,110−117.
Kohen, R., Metcalf, M. A., Khan, N., Druck, T., Huebner, K., Lachowicz, J. E.,
et al. (1996). Cloning, characterization, and chromosomal localization of a
human 5-HT6 serotonin receptor. Journal of Neurochemistry,66,47−56.
Merce-Vidal, R., Andaluz-Mataro, B., & Frigola-Constansa, J. (2003).
Preparation of N-(1H-indol-5-yl) sulfonamide derivatives with 5-HT6
receptor antagonist activity, their preparation, and their application as
medicaments for CNS diseases. WO 2003/042175 A1, Laboratorios Dr.
Esteve S.A.
Mitchell, E. S., & Neumaier, J. F. (2005). 5-HT6 receptors: A novel target for
cognitive enhancement. Pharmacology & Therapeutics,108, 320−333.
Pauwels, P. J., Tardif, S., Wurch, T., & Colpaert, F. C. (2000). Facilitation of
constitutive alpha(2A)-adrenoceptor activity by both single amino acid
mutation (Thr(373)Lys) and g(alphao) protein coexpression: Evidence for
inverse agonism. Journal of Pharmacology and Experimental Therapeutics,
292, 654−663.
Pauwels, P. J., & Wurch, T. (1998). Amino acid domains involved in constitutive
activation of G-protein-coupled receptors. Molecular Neurobiology,17,
109−135.
Purohit, A., Herrick-Davis, K., & Teitler, M. (2003). Creation, expression and
characterization of a constitutively active mutant of the human serotonin 5-
HT6 receptor. Synapse,47, 218−224.
Purohit, A., Smith, C., Herrick-Davis, K., & Teitler, M. (2005). Stable
expression of constitutively activated mutant h5HT6 and h5HT7 serotonin
receptors: Inverse agonist activity of antipsychotic drugs. Psychopharma-
cology,179, 461−469.
Reavill, C., & Rogers, D. C. (2001). The therapeutic potential of 5-HT6 receptor
antagonists. Current Opinion in Investigational Drugs,2, 104−109.
Routledge, C., Bromidge, S. M., Moss, S. F., Price, G. W., Hirst, W., Newman,
H., et al. (2000). Characterization of SB-271046: A potent, selective and
orally active 5-HT6 receptor antagonist. British Journal of Pharmacology,
130, 1606−1612.
Ruat, M.,Traiffort,E., Arrang, J. M., Tardivel-Lacombe,J., Diaz, J., Leurs, R., et al.
(1993). A novel rat serotonin(5-HT6) receptor:Molecular cloning,localization
and stimulation of cAMP accumulation. Biochemical and Biophysical
Research Communications,193,268−276.
Schoeffter, P., & Waeber, C. (1994). 5-Hydroxytryptamine receptors witha 5-HT6
receptor-like profile stimulating adenylyl cyclase activity in pig caudate
membranes. Naunyn-Schmiedeberg's Archives of Pharmacology,350,
356−360.
Sebben, M., Ansana, Y. H., Bockaert, J., & Dumuis, A. (1994). 5-HT6 receptors
positively coupled to adenylyl cyclase in striatal neurones in culture.
NeuroReport,5, 2553−2557.
Sleight, A. J., Boess, F. G., Bos, M., Levet-Trafit, B., Riemer, C., & Bourson, A.
(1998). Characterization of Ro 04-6790 and Ro 63-0563: Potent and selective
antagonists at human and rat 5-HT6 receptors. British Journal of Pharmacol-
ogy,124,556−562.
Teitler, M., Herrick-Davis, K., & Purohit, A. (2002). Constitutive activity of
G-protein coupled receptors: Emphasis on serotonin receptors. Current
Topics in Medicinal Chemistry,2,529−538.
Woolley, M. L., Marsden, C. A., & Fone, K. C. (2004). 5-ht6 receptors. Current
Drug Targets. CNS and Neurological Disorders,3,59−79.
150 G. Romero et al. / Journal of Pharmacological and Toxicological Methods 55 (2007) 144–150